diff --git a/.DS_Store b/.DS_Store
new file mode 100644
index 0000000000000000000000000000000000000000..9bb86a109357c2f056f0502b1aca097a4aa38729
Binary files /dev/null and b/.DS_Store differ
diff --git a/by-dataset/aishell4/sample_000/record.json b/by-dataset/aishell4/sample_000/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..00b35cbc297ec963d9e515836254063cf7e73c79
--- /dev/null
+++ b/by-dataset/aishell4/sample_000/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.97,
+    "samples": 143281
+  },
+  "environment_audio_duration": 5.970041751861572,
+  "environment_audio_text": "身上着火可就地打滚儿。或用厚重衣物覆盖压灭火苗儿。",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.7063,
+    "samples": 304950
+  },
+  "speaker_audio_duration": 12.706250190734863,
+  "speaker_audio_text": "器具要防止煤气泄露，使用完毕应关闭气源，就把那把那个燃气管道，和那个开关一定要关掉啊。这样不关掉很危险的。",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.7063,
+    "samples": 304950
+  },
+  "text": "冒火花发出焦糊的异味儿等情况的。应该立即切断电源开关，停止使用。第四，使用煤气",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.6813,
+    "samples": 304350
+  },
+  "speech_duration": 12.681249618530273,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__L_R004S06C01",
+  "speaker_id": "006-F",
+  "env_id": "test__L_R004S06C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 0
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_002/record.json b/by-dataset/aishell4/sample_002/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..a9f32af7f70885a6c3aaff2c8c797222bb280038
--- /dev/null
+++ b/by-dataset/aishell4/sample_002/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.63,
+    "samples": 183120
+  },
+  "environment_audio_duration": 7.630000114440918,
+  "environment_audio_text": "嗯<sil>随着呃进入冬季以后，日化是，日化，商品的这一块儿竞竞争进入了白热化的阶段<#>",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.335,
+    "samples": 152040
+  },
+  "speaker_audio_duration": 6.335000038146973,
+  "speaker_audio_text": "价签儿放错的情况下。这样的话对于销售，对于顾客购买。都会产生一种误导产生",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.335,
+    "samples": 152040
+  },
+  "text": "有一些因为家电嘛它都有型号儿有有规格儿，所以有些价签儿都属于有标错的情况，也有",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.2,
+    "samples": 148800
+  },
+  "speech_duration": 6.199999809265137,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "005-M",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 2
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_003/record.json b/by-dataset/aishell4/sample_003/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..e6d68b3fba2b4360447560ceca3cc12e5fd204c4
--- /dev/null
+++ b/by-dataset/aishell4/sample_003/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.92,
+    "samples": 118080
+  },
+  "environment_audio_duration": 4.920000076293945,
+  "environment_audio_text": "剩下的你这块儿店面的<sil>呃<sil>摆放",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.92,
+    "samples": 118081
+  },
+  "speaker_audio_duration": 4.920041561126709,
+  "speaker_audio_text": "嗯，不<->不过有的这根本就找不着售后队伍啊，或者他找也找不着很专业的",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.9201,
+    "samples": 118082
+  },
+  "text": "价签儿放错的情况下。这样的话对于销售，对于顾客购买。都会产生一种误导产生",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.335,
+    "samples": 152040
+  },
+  "speech_duration": 6.335000038146973,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "005-M",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 3
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_004/record.json b/by-dataset/aishell4/sample_004/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..4566cf592fe4d7589260e67a6f42c2ce4c0583b3
--- /dev/null
+++ b/by-dataset/aishell4/sample_004/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 9.84,
+    "samples": 236160
+  },
+  "environment_audio_duration": 9.84000015258789,
+  "environment_audio_text": "嗯，日化现在旧产品已经满足不了消费者的需求了。当地的消费者对新产品的热爱跟追求过高。",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.04,
+    "samples": 288960
+  },
+  "speaker_audio_duration": 12.039999961853027,
+  "speaker_audio_text": "所以说有因为很很多顾客，因为这个售后售后的这个问题解决不了，甚至来找我们销售员要提出退货，这个是这个也挺麻烦，嗯，是",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.04,
+    "samples": 288960
+  },
+  "text": "不不好的影响吧，这个希望这个咱超市能着重对于价签儿这块儿做一下管理",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.96,
+    "samples": 167040
+  },
+  "speech_duration": 6.960000038146973,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "005-M",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 4
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_005/record.json b/by-dataset/aishell4/sample_005/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..171fd4933ef55ce9856e973a992bf3cdf0e1f101
--- /dev/null
+++ b/by-dataset/aishell4/sample_005/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.89,
+    "samples": 309361
+  },
+  "environment_audio_duration": 12.89004135131836,
+  "environment_audio_text": "大火封门儿没有办法逃的时候，可用浸湿的毛巾，衣物堵塞门缝，发出求救信号儿，等待救援。然后这些基本上就这些等",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.6859,
+    "samples": 208461
+  },
+  "speaker_audio_duration": 8.685874938964844,
+  "speaker_audio_text": "再涂一些清凉油就行了。嗯<sil>第三呢就是使用中发现电器有冒烟儿。",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.6859,
+    "samples": 208461
+  },
+  "text": "器具要防止煤气泄露，使用完毕应关闭气源，就把那把那个燃气管道，和那个开关一定要关掉啊。这样不关掉很危险的。",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.7063,
+    "samples": 304950
+  },
+  "speech_duration": 12.706250190734863,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__L_R004S06C01",
+  "speaker_id": "006-F",
+  "env_id": "test__L_R004S06C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 5
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_007/record.json b/by-dataset/aishell4/sample_007/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4e7cd2252dd43aa8ebad662ec7d6affe7754b59
--- /dev/null
+++ b/by-dataset/aishell4/sample_007/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.9667,
+    "samples": 119200
+  },
+  "environment_audio_duration": 4.9666666984558105,
+  "environment_audio_text": "呃<sil>今天的会议主题让你们积极呃积极一点儿，",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.2,
+    "samples": 148800
+  },
+  "speaker_audio_duration": 6.199999809265137,
+  "speaker_audio_text": "有一些因为家电嘛它都有型号儿有有规格儿，所以有些价签儿都属于有标错的情况，也有",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.2,
+    "samples": 148800
+  },
+  "text": "人员我们会合理安排时，上班儿时间有培训的，有上班儿还有休息的，嗯这块儿，",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__005-M__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.6003,
+    "samples": 110408
+  },
+  "speech_duration": 4.600333213806152,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "005-M",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 7
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_008/record.json b/by-dataset/aishell4/sample_008/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..8a2e424f94896d6f7a5ec78811ace603c3fec452
--- /dev/null
+++ b/by-dataset/aishell4/sample_008/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.96,
+    "samples": 167040
+  },
+  "environment_audio_duration": 6.960000038146973,
+  "environment_audio_text": "不不好的影响吧，这个希望这个咱超市能着重对于价签儿这块儿做一下管理",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.885,
+    "samples": 213240
+  },
+  "speaker_audio_duration": 8.885000228881836,
+  "speaker_audio_text": "马上就进入双十二的时候了，嗯，那么我们日化市场就需要有新的产品推出",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.885,
+    "samples": 213240
+  },
+  "text": "那下面呃让我们日化部门说一下吧<#>",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.62,
+    "samples": 86880
+  },
+  "speech_duration": 3.619999885559082,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 8
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_009/record.json b/by-dataset/aishell4/sample_009/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..a146be9aa5ef11369f228a02eae0a496403a1aa6
--- /dev/null
+++ b/by-dataset/aishell4/sample_009/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.335,
+    "samples": 152040
+  },
+  "environment_audio_duration": 6.335000038146973,
+  "environment_audio_text": "价签儿放错的情况下。这样的话对于销售，对于顾客购买。都会产生一种误导产生",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.34,
+    "samples": 104160
+  },
+  "speaker_audio_duration": 4.340000152587891,
+  "speaker_audio_text": "同时呢还从价位上做出了理性的选择",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.34,
+    "samples": 104160
+  },
+  "text": "嗯<sil>随着呃进入冬季以后，日化是，日化，商品的这一块儿竞竞争进入了白热化的阶段<#>",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.63,
+    "samples": 183120
+  },
+  "speech_duration": 7.630000114440918,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 9
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_010/record.json b/by-dataset/aishell4/sample_010/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..0f8562f5c8706660c36c20d2e031dc971782b157
--- /dev/null
+++ b/by-dataset/aishell4/sample_010/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.745,
+    "samples": 137881
+  },
+  "environment_audio_duration": 5.745041847229004,
+  "environment_audio_text": "售后维修点儿所有的咱的产品包括维修的包括退换的质量变问题的",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.63,
+    "samples": 183120
+  },
+  "speaker_audio_duration": 7.630000114440918,
+  "speaker_audio_text": "嗯<sil>随着呃进入冬季以后，日化是，日化，商品的这一块儿竞竞争进入了白热化的阶段<#>",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.63,
+    "samples": 183120
+  },
+  "text": "嗯<sil>从是嗯双十一开始嗯，线上线下都已经就是。嗯，销售的嗯非常火爆",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 9.56,
+    "samples": 229440
+  },
+  "speech_duration": 9.5600004196167,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 10
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_011/record.json b/by-dataset/aishell4/sample_011/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..72d0e35e0173e48117ffb063d82cc68a31d7ae5b
--- /dev/null
+++ b/by-dataset/aishell4/sample_011/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 10.1,
+    "samples": 242400
+  },
+  "environment_audio_duration": 10.100000381469727,
+  "environment_audio_text": "然后穿穿越浓烟逃生时，尽量用浸湿的衣物保护住头部和身体，捂住口鼻。弯腰低姿前行。",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.7063,
+    "samples": 304950
+  },
+  "speaker_audio_duration": 12.706250190734863,
+  "speaker_audio_text": "器具要防止煤气泄露，使用完毕应关闭气源，就把那把那个燃气管道，和那个开关一定要关掉啊。这样不关掉很危险的。",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.7063,
+    "samples": 304950
+  },
+  "text": "第五呢不要在公共场合炫耀自家的钱财如何富有",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.8306,
+    "samples": 139934
+  },
+  "speech_duration": 5.830583095550537,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__L_R004S06C01",
+  "speaker_id": "006-F",
+  "env_id": "test__L_R004S06C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 11
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_016/record.json b/by-dataset/aishell4/sample_016/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..3d2df0138981e7fee9ea02c3acd1bd0733697f1c
--- /dev/null
+++ b/by-dataset/aishell4/sample_016/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.29,
+    "samples": 318960
+  },
+  "environment_audio_duration": 13.289999961853027,
+  "environment_audio_text": "正常使用经常检查煤气，燃气开关用具，发现燃气泄漏，迅速要关闭阀门儿，打开门窗也不要去。触动电器开关和玩儿明火",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.6859,
+    "samples": 208461
+  },
+  "speaker_audio_duration": 8.685874938964844,
+  "speaker_audio_text": "再涂一些清凉油就行了。嗯<sil>第三呢就是使用中发现电器有冒烟儿。",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.6859,
+    "samples": 208461
+  },
+  "text": "父母，他只是担当何。嗯父母是担当何种部门儿要职，又在何处购了新房，有几部车等等，这些都容易被坏人利用后。踩点儿行骗。",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 14.2414,
+    "samples": 341793
+  },
+  "speech_duration": 14.241374969482422,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__L_R004S06C01",
+  "speaker_id": "006-F",
+  "env_id": "test__L_R004S06C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 16
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_017/record.json b/by-dataset/aishell4/sample_017/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..d586f8bd56bd5e2b823272fb296ed2b732cc93e5
--- /dev/null
+++ b/by-dataset/aishell4/sample_017/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.96,
+    "samples": 167040
+  },
+  "environment_audio_duration": 6.960000038146973,
+  "environment_audio_text": "不不好的影响吧，这个希望这个咱超市能着重对于价签儿这块儿做一下管理",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.78,
+    "samples": 138720
+  },
+  "speaker_audio_duration": 5.78000020980835,
+  "speaker_audio_text": "因为他们觉着嗯<sil>这么多人去买，肯定是肯定是好",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.78,
+    "samples": 138720
+  },
+  "text": "嗯<sil>只有那些品质好效果好消费者选择产品的成功率才会更大。",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.095,
+    "samples": 194280
+  },
+  "speech_duration": 8.095000267028809,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 17
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_018/record.json b/by-dataset/aishell4/sample_018/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..2b6ecb1fa54cfa345a69e7757ca61afda4e31f8a
--- /dev/null
+++ b/by-dataset/aishell4/sample_018/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.92,
+    "samples": 118080
+  },
+  "environment_audio_duration": 4.920000076293945,
+  "environment_audio_text": "剩下的你这块儿店面的<sil>呃<sil>摆放",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.27,
+    "samples": 150480
+  },
+  "speaker_audio_duration": 6.269999980926514,
+  "speaker_audio_text": "而且朋友之间也会相互介绍，他们就比较放心购买。所以在我们超市在",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.27,
+    "samples": 150480
+  },
+  "text": "嗯美容美丽的自信，他们才会放心的去买",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.63,
+    "samples": 111119
+  },
+  "speech_duration": 4.629958152770996,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 18
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_019/record.json b/by-dataset/aishell4/sample_019/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..cd22db5116fd7d61795a710f3b4f683ee35cbda2
--- /dev/null
+++ b/by-dataset/aishell4/sample_019/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.365,
+    "samples": 200761
+  },
+  "environment_audio_duration": 8.365041732788086,
+  "environment_audio_text": "高档产品多一点儿，还是低档产品多一点儿，嗯假如就是说咱不是都有品牌嘛，像欧莱雅啦，还有一些，嗯<sil>",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.095,
+    "samples": 194280
+  },
+  "speaker_audio_duration": 8.095000267028809,
+  "speaker_audio_text": "嗯<sil>只有那些品质好效果好消费者选择产品的成功率才会更大。",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.095,
+    "samples": 194280
+  },
+  "text": "产品知名度高，虽然在没有真正体验过产品，但是在当地知名度高，在当地广告儿铺天盖地的影响力",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 11.605,
+    "samples": 278520
+  },
+  "speech_duration": 11.604999542236328,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 19
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_020/record.json b/by-dataset/aishell4/sample_020/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..beca821b88e3bf63f2e5d2c38eeaf0aa66cf205e
--- /dev/null
+++ b/by-dataset/aishell4/sample_020/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 14.64,
+    "samples": 351361
+  },
+  "environment_audio_duration": 14.64004135131836,
+  "environment_audio_text": "然后教育儿童不玩儿火的时候，要将打火机火柴放到儿童拿不到的地方，再有常识，不占用阻塞或者封闭安全出口儿。疏散通道和消防车通道不妨碍。",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.8306,
+    "samples": 139934
+  },
+  "speaker_audio_duration": 5.830583095550537,
+  "speaker_audio_text": "第五呢不要在公共场合炫耀自家的钱财如何富有",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.8306,
+    "samples": 139935
+  },
+  "text": "啊这个这个是有实例的啊，我们邻居就是有，有这种情况，他家已经被盗过很多次了。啊",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.0508,
+    "samples": 193219
+  },
+  "speech_duration": 8.05079174041748,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__L_R004S06C01",
+  "speaker_id": "006-F",
+  "env_id": "test__L_R004S06C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 20
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_021/record.json b/by-dataset/aishell4/sample_021/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..09518a499855f5d2d22a30bea4514faebe294e64
--- /dev/null
+++ b/by-dataset/aishell4/sample_021/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.8877,
+    "samples": 141306
+  },
+  "environment_audio_duration": 5.887750148773193,
+  "environment_audio_text": "或者是主要就还是它是以哪个年龄段儿的，咱这个<sil>有统计吗，是",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.63,
+    "samples": 111119
+  },
+  "speaker_audio_duration": 4.629958152770996,
+  "speaker_audio_text": "嗯美容美丽的自信，他们才会放心的去买",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.63,
+    "samples": 111120
+  },
+  "text": "因为他们觉着嗯<sil>这么多人去买，肯定是肯定是好",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.78,
+    "samples": 138720
+  },
+  "speech_duration": 5.78000020980835,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 21
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_026/record.json b/by-dataset/aishell4/sample_026/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..9810025fe058d143f33af06ded41243eeb69211d
--- /dev/null
+++ b/by-dataset/aishell4/sample_026/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.365,
+    "samples": 200761
+  },
+  "environment_audio_duration": 8.365041732788086,
+  "environment_audio_text": "高档产品多一点儿，还是低档产品多一点儿，嗯假如就是说咱不是都有品牌嘛，像欧莱雅啦，还有一些，嗯<sil>",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.095,
+    "samples": 122280
+  },
+  "speaker_audio_duration": 5.09499979019165,
+  "speaker_audio_text": "现在我<->根据我们超市现在的情况来看，大部分都是普通的消费者",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.095,
+    "samples": 122280
+  },
+  "text": "嗯，就是就是产生消费的欲望",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.38,
+    "samples": 81120
+  },
+  "speech_duration": 3.380000114440918,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 26
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_027/record.json b/by-dataset/aishell4/sample_027/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..c9dd21881dee3022aea7e345c32257e62018f938
--- /dev/null
+++ b/by-dataset/aishell4/sample_027/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.2352,
+    "samples": 149646
+  },
+  "environment_audio_duration": 6.235249996185303,
+  "environment_audio_text": "就是像四十岁到六十岁就买一些家庭的一个护肤性的，包括洗发水儿啦洗发膏儿啦",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.395,
+    "samples": 201481
+  },
+  "speaker_audio_duration": 8.395041465759277,
+  "speaker_audio_text": "进货这，嗯这一方面呢应该就是就是选择范围要嗯嗯扩大一下",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.3951,
+    "samples": 201482
+  },
+  "text": "嗯<sil>还有产品的一些陈列方面。嗯，就是",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.79,
+    "samples": 138960
+  },
+  "speech_duration": 5.789999961853027,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 27
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_028/record.json b/by-dataset/aishell4/sample_028/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..20cd4fc633c64e3ef8b1dfd56c2712255e08272e
--- /dev/null
+++ b/by-dataset/aishell4/sample_028/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.365,
+    "samples": 200761
+  },
+  "environment_audio_duration": 8.365041732788086,
+  "environment_audio_text": "高档产品多一点儿，还是低档产品多一点儿，嗯假如就是说咱不是都有品牌嘛，像欧莱雅啦，还有一些，嗯<sil>",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.265,
+    "samples": 150360
+  },
+  "speaker_audio_duration": 6.264999866485596,
+  "speaker_audio_text": "嗯<sil>应该是属于就是中档一嗯一些的日用品比较好",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.265,
+    "samples": 150360
+  },
+  "text": "还有什么嗯染烫用品呀，嗯应该就是更趋于合理一些",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__S_R003S01C01__002-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.25,
+    "samples": 174000
+  },
+  "speech_duration": 7.25,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__S_R003S01C01",
+  "speaker_id": "002-F",
+  "env_id": "test__S_R003S01C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 28
+}
\ No newline at end of file
diff --git a/by-dataset/aishell4/sample_029/record.json b/by-dataset/aishell4/sample_029/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbb2f5282803f4f4cf29f9c2d1b5073f35bc3164
--- /dev/null
+++ b/by-dataset/aishell4/sample_029/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.005,
+    "samples": 312121
+  },
+  "environment_audio_duration": 13.00504207611084,
+  "environment_audio_text": "然后学校和单位定期组织逃生疏散演练，进入公共场所的时候，需要注意观察安全出口儿和疏散通道，记住疏散方向",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.8306,
+    "samples": 139934
+  },
+  "speaker_audio_duration": 5.830583095550537,
+  "speaker_audio_text": "第五呢不要在公共场合炫耀自家的钱财如何富有",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.8306,
+    "samples": 139935
+  },
+  "text": "也不要在同学，朋友及陌生人面前显富。吹嘘自家财产，炫耀自己的衣服鞋帽是名牌儿。是花多少钱买的等等，",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "aishell4__test__L_R004S06C01__006-F__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.2363,
+    "samples": 317672
+  },
+  "speech_duration": 13.236332893371582,
+  "language": "zh",
+  "dataset": "aishell4",
+  "conversation_id": "test__L_R004S06C01",
+  "speaker_id": "006-F",
+  "env_id": "test__L_R004S06C01",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 29
+}
\ No newline at end of file
diff --git a/by-dataset/alimeeting/sample_029/record.json b/by-dataset/alimeeting/sample_029/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..cde9b7ef1ae059aef6e1ea5909178ec496603d04
--- /dev/null
+++ b/by-dataset/alimeeting/sample_029/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "alimeeting__R1021_M4078_MS104__F_SPK4086__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.32,
+    "samples": 103680
+  },
+  "environment_audio_duration": 4.320000171661377,
+  "environment_audio_text": "很多客户，会反映啊，就最近接到的一些投诉，就是说。",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "alimeeting__R1021_M4078_MS104__F_SPK4086__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.56,
+    "samples": 85440
+  },
+  "speaker_audio_duration": 3.559999942779541,
+  "speaker_audio_text": "呃一个，呃，服务服务流程。",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "alimeeting__R1021_M4078_MS104__F_SPK4086__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.56,
+    "samples": 85440
+  },
+  "text": "呃，像我们有专门的那种配送车辆，公司。",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "alimeeting__R1021_M4078_MS104__F_SPK4086__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.33,
+    "samples": 79920
+  },
+  "speech_duration": 3.3299999237060547,
+  "language": "zh",
+  "dataset": "alimeeting",
+  "conversation_id": "R1021_M4078_MS104",
+  "speaker_id": "F_SPK4086",
+  "env_id": "R1021_M4078_MS104_ch0",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 29
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_000/record.json b/by-dataset/chime6/sample_000/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..f2d49400131deab3bc7b93f410cbb1cfda300cd3
--- /dev/null
+++ b/by-dataset/chime6/sample_000/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.6,
+    "samples": 86400
+  },
+  "environment_audio_duration": 3.5999999046325684,
+  "environment_audio_text": "But, I I mean it it's okay if you're just using water like if you're rinsing yourself that's cool but",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "speaker_audio_duration": 3.2100000381469727,
+  "speaker_audio_text": "Just went all out. I think yes.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "text": "Yeah, no cuz uh, well cuz I I took that that course and having to like speak something for a recording like I understand what the people who are doing this have to go through",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.93,
+    "samples": 142320
+  },
+  "speech_duration": 5.929999828338623,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 0
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_002/record.json b/by-dataset/chime6/sample_002/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..7024d7dc4c5d03a30fdf222cb862dd100d17a210
--- /dev/null
+++ b/by-dataset/chime6/sample_002/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.64,
+    "samples": 87360
+  },
+  "environment_audio_duration": 3.640000104904175,
+  "environment_audio_text": "I got to what most people would cur call a menial level.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "speaker_audio_duration": 3.2100000381469727,
+  "speaker_audio_text": "Just went all out. I think yes.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "text": "The the, it's bad like I've read that it's bad to take a shower every single day.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.5,
+    "samples": 84000
+  },
+  "speech_duration": 3.5,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 2
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_003/record.json b/by-dataset/chime6/sample_003/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..19397ef53b8e7274d5bc78a2aba9ec0eec57bf5f
--- /dev/null
+++ b/by-dataset/chime6/sample_003/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.48,
+    "samples": 83520
+  },
+  "environment_audio_duration": 3.4800000190734863,
+  "environment_audio_text": "I envy you I got- I gotta be clean.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.39,
+    "samples": 81360
+  },
+  "speaker_audio_duration": 3.390000104904175,
+  "speaker_audio_text": "Why didn't we take pictures of that? I would've loved to see the stuff today.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.39,
+    "samples": 81360
+  },
+  "text": "Well your clothes will give you the best shower that you ever had which is what you wash out you take it off all the crap that's on the inside of your shirt. Change your outfit once a day.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.66,
+    "samples": 159840
+  },
+  "speech_duration": 6.659999847412109,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 3
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_004/record.json b/by-dataset/chime6/sample_004/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..b4e60567fbddfccfdd3c8194ed3912467703c387
--- /dev/null
+++ b/by-dataset/chime6/sample_004/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.6,
+    "samples": 86400
+  },
+  "environment_audio_duration": 3.5999999046325684,
+  "environment_audio_text": "But, I I mean it it's okay if you're just using water like if you're rinsing yourself that's cool but",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.51,
+    "samples": 108240
+  },
+  "speaker_audio_duration": 4.510000228881836,
+  "speaker_audio_text": "Dude, I think the the ten percent of your day should just be kept to like keeping the forest out there.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.51,
+    "samples": 108240
+  },
+  "text": "So then, and I trust that he's telling you I'm gonna learn and you're gonna sit.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.36,
+    "samples": 80640
+  },
+  "speech_duration": 3.359999895095825,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 4
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_005/record.json b/by-dataset/chime6/sample_005/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..245abba6d9b8b88c51c7e120091c1cf9b9dd5d57
--- /dev/null
+++ b/by-dataset/chime6/sample_005/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.25,
+    "samples": 78000
+  },
+  "environment_audio_duration": 3.25,
+  "environment_audio_text": "Yeah dude, I- I'm not doing you a service",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.93,
+    "samples": 142320
+  },
+  "speaker_audio_duration": 5.929999828338623,
+  "speaker_audio_text": "Yeah, no cuz uh, well cuz I I took that that course and having to like speak something for a recording like I understand what the people who are doing this have to go through",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.93,
+    "samples": 142320
+  },
+  "text": "I'm sure the paint on this one is a lot better than the one you're getting from, cuz this is like the wood itself is dyed rather than painted.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.69,
+    "samples": 136560
+  },
+  "speech_duration": 5.690000057220459,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 5
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_007/record.json b/by-dataset/chime6/sample_007/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..148c98ae198b97125977f3ceea5eab1ac508a18a
--- /dev/null
+++ b/by-dataset/chime6/sample_007/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.47,
+    "samples": 83280
+  },
+  "environment_audio_duration": 3.4700000286102295,
+  "environment_audio_text": "To- to our surprise it was more dust that came out of it than",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.24,
+    "samples": 101760
+  },
+  "speaker_audio_duration": 4.239999771118164,
+  "speaker_audio_text": "You just need a moss in there to uh reuptake the the crap uh nitrogen in the water.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.24,
+    "samples": 101760
+  },
+  "text": "Just went all out. I think yes.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "speech_duration": 3.2100000381469727,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 7
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_008/record.json b/by-dataset/chime6/sample_008/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..5f0bdf4053a54264cedbe2f10af8a3b84371257d
--- /dev/null
+++ b/by-dataset/chime6/sample_008/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.02,
+    "samples": 72480
+  },
+  "environment_audio_duration": 3.0199999809265137,
+  "environment_audio_text": "Really? I've com- completely forgotten the story, but that's awesome.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "speaker_audio_duration": 3.2100000381469727,
+  "speaker_audio_text": "Just went all out. I think yes.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "text": "Why didn't we take pictures of that? I would've loved to see the stuff today.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.39,
+    "samples": 81360
+  },
+  "speech_duration": 3.390000104904175,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 8
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_009/record.json b/by-dataset/chime6/sample_009/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..1e19dfc3798d536c2ddf8b7fa53aba7b7f11f290
--- /dev/null
+++ b/by-dataset/chime6/sample_009/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.89,
+    "samples": 117360
+  },
+  "environment_audio_duration": 4.889999866485596,
+  "environment_audio_text": "Something to do with the height here, it's like all these guys are all mossy.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.34,
+    "samples": 104160
+  },
+  "speaker_audio_duration": 4.340000152587891,
+  "speaker_audio_text": "Not not not n- not the ones of the specific par- participants here. Cuz then",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.34,
+    "samples": 104160
+  },
+  "text": "You just need a moss in there to uh reuptake the the crap uh nitrogen in the water.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.24,
+    "samples": 101760
+  },
+  "speech_duration": 4.239999771118164,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 9
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_010/record.json b/by-dataset/chime6/sample_010/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..1e0e789f4c12114fc55d2708b64d5d190aa9bb7f
--- /dev/null
+++ b/by-dataset/chime6/sample_010/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.62,
+    "samples": 86880
+  },
+  "environment_audio_duration": 3.619999885559082,
+  "environment_audio_text": "Well, I mean, when you're going with your girlfriend, it just sounds like a good deal",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.24,
+    "samples": 101760
+  },
+  "speaker_audio_duration": 4.239999771118164,
+  "speaker_audio_text": "You just need a moss in there to uh reuptake the the crap uh nitrogen in the water.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.24,
+    "samples": 101760
+  },
+  "text": "except for the fact that we can't do it on like",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.04,
+    "samples": 72960
+  },
+  "speech_duration": 3.0399999618530273,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 10
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_011/record.json b/by-dataset/chime6/sample_011/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..6ab327dc6ed4f53824b416b01ab59d940c540469
--- /dev/null
+++ b/by-dataset/chime6/sample_011/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.58,
+    "samples": 109920
+  },
+  "environment_audio_duration": 4.579999923706055,
+  "environment_audio_text": "Two two have died but there's two more. One's already flowering an- or blooming. The other one's comin' up.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.5,
+    "samples": 84000
+  },
+  "speaker_audio_duration": 3.5,
+  "speaker_audio_text": "I apologize in advance if my speaking is very fast and hard to transcribe.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.5,
+    "samples": 84000
+  },
+  "text": "Yo, we should've we nev- we should've asked the question though. What if you need to use the washroom?",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.27,
+    "samples": 78480
+  },
+  "speech_duration": 3.2699999809265137,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 11
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_016/record.json b/by-dataset/chime6/sample_016/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..87bc1b8efb9be6a93cb2adeeddfbd2a171bf4d53
--- /dev/null
+++ b/by-dataset/chime6/sample_016/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.31,
+    "samples": 79440
+  },
+  "environment_audio_duration": 3.309999942779541,
+  "environment_audio_text": "Dude there seriously has to be something better than this.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.02,
+    "samples": 144480
+  },
+  "speaker_audio_duration": 6.019999980926514,
+  "speaker_audio_text": "I'm doing uh like that dinner thing slash lunch thing right now so you're being recorded but what's up?",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.02,
+    "samples": 144480
+  },
+  "text": "Yo do you think that pollu- is that fire pollution or is that just pollution?",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.87,
+    "samples": 92880
+  },
+  "speech_duration": 3.869999885559082,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 16
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_017/record.json b/by-dataset/chime6/sample_017/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..9e513170d6e6f063a8c0468b43c34ea1e6a6efc4
--- /dev/null
+++ b/by-dataset/chime6/sample_017/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.25,
+    "samples": 78000
+  },
+  "environment_audio_duration": 3.25,
+  "environment_audio_text": "I I've heard that Himalayan Himalayan salt has more radiation in it.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.66,
+    "samples": 87840
+  },
+  "speaker_audio_duration": 3.6600000858306885,
+  "speaker_audio_text": "I mean you could turn your balcony into nothing but shelving units and just grow.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.66,
+    "samples": 87840
+  },
+  "text": "I'm doing uh like that dinner thing slash lunch thing right now so you're being recorded but what's up?",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.02,
+    "samples": 144480
+  },
+  "speech_duration": 6.019999980926514,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 17
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_018/record.json b/by-dataset/chime6/sample_018/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..9d977e1ec319fc5f9ceb19b0d6e716ad530fd0a1
--- /dev/null
+++ b/by-dataset/chime6/sample_018/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.28,
+    "samples": 78720
+  },
+  "environment_audio_duration": 3.2799999713897705,
+  "environment_audio_text": "Or it ha- has the possibility to have more. I have no idea if it's true or not.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.66,
+    "samples": 87840
+  },
+  "speaker_audio_duration": 3.6600000858306885,
+  "speaker_audio_text": "I mean you could turn your balcony into nothing but shelving units and just grow.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.66,
+    "samples": 87840
+  },
+  "text": "Cuz with the with like salt water taken from like the ocean like it has dissolved plastics and like plastic particles in there.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.98,
+    "samples": 167520
+  },
+  "speech_duration": 6.980000019073486,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 18
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_019/record.json b/by-dataset/chime6/sample_019/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..1bc466d02b12e88c562a8a4845c77e6cc7ad8f18
--- /dev/null
+++ b/by-dataset/chime6/sample_019/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 9.37,
+    "samples": 224880
+  },
+  "environment_audio_duration": 9.369999885559082,
+  "environment_audio_text": "I think the main thing is is that we have to go, like we have to look back and be like \"Okay, are we having so much of this that it's gonna affect our",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8,
+    "samples": 91200
+  },
+  "speaker_audio_duration": 3.799999952316284,
+  "speaker_audio_text": "Would you, how, do you, do you guys just start rolling around on the ground until it stopped or",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8,
+    "samples": 91200
+  },
+  "text": "Yo I got brown sugar. You should put some brown sugar in there.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.21,
+    "samples": 77040
+  },
+  "speech_duration": 3.2100000381469727,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 19
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_020/record.json b/by-dataset/chime6/sample_020/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..024a81854a8cfcb927fc6c4c4b6ab23bcec41c60
--- /dev/null
+++ b/by-dataset/chime6/sample_020/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.47,
+    "samples": 107280
+  },
+  "environment_audio_duration": 4.46999979019165,
+  "environment_audio_text": "Th- they say cuz it's it comes from different regions on the planet that have like higher radioactivity.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8,
+    "samples": 91200
+  },
+  "speaker_audio_duration": 3.799999952316284,
+  "speaker_audio_text": "Would you, how, do you, do you guys just start rolling around on the ground until it stopped or",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8,
+    "samples": 91200
+  },
+  "text": "No I will, they'll still cut the like, I'm sure the acidity will take will like open up like the chili flakes and make it super s- super",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.49,
+    "samples": 155760
+  },
+  "speech_duration": 6.489999771118164,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 20
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_021/record.json b/by-dataset/chime6/sample_021/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..c9d2c2f8611d4273ea90d1536cf3fcac65e34433
--- /dev/null
+++ b/by-dataset/chime6/sample_021/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.92,
+    "samples": 118080
+  },
+  "environment_audio_duration": 4.920000076293945,
+  "environment_audio_text": "This man offers his estate to people who can survive the sauce. You must've been in agony.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.66,
+    "samples": 135840
+  },
+  "speaker_audio_duration": 5.659999847412109,
+  "speaker_audio_text": "cut across the top of the potato and squeeze the sides. Add a dollop of sour cream on your, and your favorite topping.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.66,
+    "samples": 135840
+  },
+  "text": "Would you, how, do you, do you guys just start rolling around on the ground until it stopped or",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8,
+    "samples": 91200
+  },
+  "speech_duration": 3.799999952316284,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 21
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_026/record.json b/by-dataset/chime6/sample_026/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..ad276c7343dfb2d707f6e471cb526207173c4155
--- /dev/null
+++ b/by-dataset/chime6/sample_026/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.02,
+    "samples": 96480
+  },
+  "environment_audio_duration": 4.019999980926514,
+  "environment_audio_text": "Give it back. It's like \"Here you go, it's all wet.\"",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.66,
+    "samples": 87840
+  },
+  "speaker_audio_duration": 3.6600000858306885,
+  "speaker_audio_text": "I mean you could turn your balcony into nothing but shelving units and just grow.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.66,
+    "samples": 87840
+  },
+  "text": "cut across the top of the potato and squeeze the sides. Add a dollop of sour cream on your, and your favorite topping.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.66,
+    "samples": 135840
+  },
+  "speech_duration": 5.659999847412109,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 26
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_027/record.json b/by-dataset/chime6/sample_027/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..5f93fb0ab2271e46973a2b9e634f3c9fe80946a4
--- /dev/null
+++ b/by-dataset/chime6/sample_027/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.14,
+    "samples": 99360
+  },
+  "environment_audio_duration": 4.139999866485596,
+  "environment_audio_text": "Yeah all the memes say that it's Ubisoft is I mean bad but does EA have a better",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.22,
+    "samples": 149280
+  },
+  "speaker_audio_duration": 6.21999979019165,
+  "speaker_audio_text": "Yo when I went worked at Taco they are their fridge would repeatedly say like negative thirty Fahrenheit and shit like that.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 6.22,
+    "samples": 149280
+  },
+  "text": "It's too bad you guys don't own a cat. I would love to play with a cat right now.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.23,
+    "samples": 77520
+  },
+  "speech_duration": 3.2300000190734863,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 27
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_028/record.json b/by-dataset/chime6/sample_028/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..648683374d3e711982b98d025b3dc3503b7fb877
--- /dev/null
+++ b/by-dataset/chime6/sample_028/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.23,
+    "samples": 77520
+  },
+  "environment_audio_duration": 3.2300000190734863,
+  "environment_audio_text": "Check this out. Okay no, I'm gonna make a game for you guys, okay guys?",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.32,
+    "samples": 79680
+  },
+  "speaker_audio_duration": 3.319999933242798,
+  "speaker_audio_text": "Oh no not not like not like grow, no. Like a cooking pot.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.32,
+    "samples": 79680
+  },
+  "text": "Yeah, this one right here you can see it, but I mean I don't think you can make out the TV screen.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.49,
+    "samples": 107760
+  },
+  "speech_duration": 4.489999771118164,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 28
+}
\ No newline at end of file
diff --git a/by-dataset/chime6/sample_029/record.json b/by-dataset/chime6/sample_029/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..8380d5276046644425eccd651a81cf57a9bd81af
--- /dev/null
+++ b/by-dataset/chime6/sample_029/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.15,
+    "samples": 75600
+  },
+  "environment_audio_duration": 3.1500000953674316,
+  "environment_audio_text": "Rick when he was a pickle.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.49,
+    "samples": 107760
+  },
+  "speaker_audio_duration": 4.489999771118164,
+  "speaker_audio_text": "Yeah, this one right here you can see it, but I mean I don't think you can make out the TV screen.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "chime6__train__S12__P35__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.49,
+    "samples": 107760
+  },
+  "text": "Yes I have uh I have my pope-mobile.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "chime6__train__S12__P35__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.5,
+    "samples": 84000
+  },
+  "speech_duration": 3.5,
+  "language": "en",
+  "dataset": "chime6",
+  "conversation_id": "train__S12",
+  "speaker_id": "P35",
+  "env_id": "S12",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 29
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_000/record.json b/by-dataset/dds/sample_000/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..b8cf8733ea5a2f1878fae99fc20c17665b48a625
--- /dev/null
+++ b/by-dataset/dds/sample_000/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Uber__ch1__m3__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8,
+    "samples": 91199
+  },
+  "environment_audio_duration": 3.7999582290649414,
+  "environment_audio_text": "Alice was beginning to get very tired of sitting by her sister on the bank,",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Uber__ch1__m3__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.5829,
+    "samples": 133989
+  },
+  "speaker_audio_duration": 5.582874774932861,
+  "speaker_audio_text": "by slouching about the banks and blackmailing weakminded noodles in this way.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Uber__ch1__m3__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.5829,
+    "samples": 133989
+  },
+  "text": "my heart sank, as the saying is, into my boots;",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Uber__ch1__m3__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.5029,
+    "samples": 108069
+  },
+  "speech_duration": 4.502874851226807,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Uber__ch1",
+  "speaker_id": "m3",
+  "env_id": "dds__office2__Uber__ch1",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 0
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_001/record.json b/by-dataset/dds/sample_001/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..6e10c84b38f8d8c403a11b1fff29ea0e3b3c7241
--- /dev/null
+++ b/by-dataset/dds/sample_001/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch4__m1__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.653,
+    "samples": 135671
+  },
+  "environment_audio_duration": 5.652958393096924,
+  "environment_audio_text": "falling from four frosted half globes set in the scrollwork of the ceiling.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch4__m1__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.901,
+    "samples": 117624
+  },
+  "speaker_audio_duration": 4.901000022888184,
+  "speaker_audio_text": "for it flashed across her mind that she had never before seen a rabbit with either a waistcoat-pocket,",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__iPad__ch4__m1__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.901,
+    "samples": 117624
+  },
+  "text": "The heat was sweltering, and the men grumbled fiercely over their work.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__iPad__ch4__m1__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.84,
+    "samples": 92160
+  },
+  "speech_duration": 3.8399999141693115,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__iPad__ch4",
+  "speaker_id": "m1",
+  "env_id": "dds__office2__iPad__ch4",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 1
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_007/record.json b/by-dataset/dds/sample_007/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..b024a37d7cae609ccd3fea88268e7cf07793e895
--- /dev/null
+++ b/by-dataset/dds/sample_007/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch5__m1__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.3502,
+    "samples": 104406
+  },
+  "environment_audio_duration": 4.350249767303467,
+  "environment_audio_text": "and even beyond, where the paddocks were, and the berry patches.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch5__m1__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.5265,
+    "samples": 60636
+  },
+  "speaker_audio_duration": 2.5264999866485596,
+  "speaker_audio_text": "and the world turned giddily before my eyes,",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__iPad__ch5__m1__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.5265,
+    "samples": 60636
+  },
+  "text": "and from the first look onward, I hated the very thought of Treasure Island.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__iPad__ch5__m1__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.5297,
+    "samples": 108714
+  },
+  "speech_duration": 4.529749870300293,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__iPad__ch5",
+  "speaker_id": "m1",
+  "env_id": "dds__office2__iPad__ch5",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 7
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_008/record.json b/by-dataset/dds/sample_008/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..53e0956ab51db7834f7c092a8217e8f4ae4eb713
--- /dev/null
+++ b/by-dataset/dds/sample_008/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch3__m1__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.6817,
+    "samples": 64361
+  },
+  "environment_audio_duration": 2.681708335876465,
+  "environment_audio_text": "and here he had lived the four years of his life.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch3__m1__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.84,
+    "samples": 92160
+  },
+  "speaker_audio_duration": 3.8399999141693115,
+  "speaker_audio_text": "The heat was sweltering, and the men grumbled fiercely over their work.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Marantz__ch3__m1__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.84,
+    "samples": 92160
+  },
+  "text": "my heart sank, as the saying is, into my boots;",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Marantz__ch3__m1__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.0196,
+    "samples": 96471
+  },
+  "speech_duration": 4.019625186920166,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Marantz__ch3",
+  "speaker_id": "m1",
+  "env_id": "dds__office2__Marantz__ch3",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 8
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_009/record.json b/by-dataset/dds/sample_009/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..229187d3d2ea939d92d3d412a1cff41bf66d3052
--- /dev/null
+++ b/by-dataset/dds/sample_009/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch5__f8__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.5425,
+    "samples": 85019
+  },
+  "environment_audio_duration": 3.5424582958221436,
+  "environment_audio_text": "He plunged into the swimming tank or went hunting with the Judge's sons;",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch5__f8__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.3851,
+    "samples": 177242
+  },
+  "speaker_audio_duration": 7.385083198547363,
+  "speaker_audio_text": "We said we hadn't given the matter sufficient consideration as yet to enable us to arrive at a definite conclusion on that point,",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Marantz__ch5__f8__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.3851,
+    "samples": 177242
+  },
+  "text": "but the very sight of the island had relaxed the cords of discipline.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Marantz__ch5__f8__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.5692,
+    "samples": 85662
+  },
+  "speech_duration": 3.5692501068115234,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Marantz__ch5",
+  "speaker_id": "f8",
+  "env_id": "dds__office2__Marantz__ch5",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 9
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_010/record.json b/by-dataset/dds/sample_010/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..59e6e0e68c415e44f26ff7f913f58053e9d2952a
--- /dev/null
+++ b/by-dataset/dds/sample_010/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Uber__ch2__m2__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.7865,
+    "samples": 90876
+  },
+  "environment_audio_duration": 3.7864999771118164,
+  "environment_audio_text": "We stopped under the willows by Kempton Park, and lunched.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Uber__ch2__m2__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.6827,
+    "samples": 64386
+  },
+  "speaker_audio_duration": 2.6827499866485596,
+  "speaker_audio_text": "above all in the morning, on an empty stomach.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Uber__ch2__m2__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.6827,
+    "samples": 64386
+  },
+  "text": "out-topping the others｡ｪsome singly, some in clumps;",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Uber__ch2__m2__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.6739,
+    "samples": 88173
+  },
+  "speech_duration": 3.673875093460083,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Uber__ch2",
+  "speaker_id": "m2",
+  "env_id": "dds__office2__Uber__ch2",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 10
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_017/record.json b/by-dataset/dds/sample_017/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..003934a17c5df98aceb8c796adf5c9742cd38f09
--- /dev/null
+++ b/by-dataset/dds/sample_017/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch6__f4__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.9912,
+    "samples": 95790
+  },
+  "environment_audio_duration": 3.9912500381469727,
+  "environment_audio_text": "Contrived at the rear of the dining room, a double door opened,",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch6__f4__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.7014,
+    "samples": 64833
+  },
+  "speaker_audio_duration": 2.7013750076293945,
+  "speaker_audio_text": "Harris said that if it was a duty it ought to be done,",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Marantz__ch6__f4__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.7014,
+    "samples": 64833
+  },
+  "text": "At the rear things were on even a more spacious scale than at the front.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Marantz__ch6__f4__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.8476,
+    "samples": 92343
+  },
+  "speech_duration": 3.8476250171661377,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Marantz__ch6",
+  "speaker_id": "f4",
+  "env_id": "dds__office2__Marantz__ch6",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 17
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_019/record.json b/by-dataset/dds/sample_019/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..fd5a066a27809e2d5c889cf33243c725d0296801
--- /dev/null
+++ b/by-dataset/dds/sample_019/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch4__m5__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.2547,
+    "samples": 78113
+  },
+  "environment_audio_duration": 3.2547082901000977,
+  "environment_audio_text": "We stopped under the willows by Kempton Park, and lunched.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Marantz__ch4__m5__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.943,
+    "samples": 70631
+  },
+  "speaker_audio_duration": 2.942958354949951,
+  "speaker_audio_text": "They're my sole remaining ties with dry land.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Marantz__ch4__m5__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.943,
+    "samples": 70631
+  },
+  "text": "and was just in time to see it pop down a large rabbit-hole under the hedge.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Marantz__ch4__m5__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.9613,
+    "samples": 119072
+  },
+  "speech_duration": 4.961333274841309,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Marantz__ch4",
+  "speaker_id": "m5",
+  "env_id": "dds__office2__Marantz__ch4",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 19
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_021/record.json b/by-dataset/dds/sample_021/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..477e46cbbb952b81fb92861967c980d3a31d3e90
--- /dev/null
+++ b/by-dataset/dds/sample_021/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch3__f9__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.4949,
+    "samples": 83877
+  },
+  "environment_audio_duration": 3.494874954223633,
+  "environment_audio_text": "we would, without further hesitation, believe it.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch3__f9__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.3366,
+    "samples": 80078
+  },
+  "speaker_audio_duration": 3.336583375930786,
+  "speaker_audio_text": "She took down a jar from one of the shelves as she passed:",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__iPad__ch3__f9__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.3366,
+    "samples": 80078
+  },
+  "text": "Grey-coloured woods covered a large part of the surface.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__iPad__ch3__f9__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.1759,
+    "samples": 76221
+  },
+  "speech_duration": 3.175874948501587,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__iPad__ch3",
+  "speaker_id": "f9",
+  "env_id": "dds__office2__iPad__ch3",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 21
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_026/record.json b/by-dataset/dds/sample_026/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..8439505af9aabf36fc31a544c5d8c4585aee71fc
--- /dev/null
+++ b/by-dataset/dds/sample_026/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch6__m1__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 2.7257,
+    "samples": 65417
+  },
+  "environment_audio_duration": 2.725708246231079,
+  "environment_audio_text": "｡ｰI must be getting somewhere near the centre of the earth.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__iPad__ch6__m1__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.7817,
+    "samples": 90761
+  },
+  "speaker_audio_duration": 3.781708240509033,
+  "speaker_audio_text": "for up to that day the men had gone briskly and willingly about their business;",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__iPad__ch6__m1__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.7817,
+    "samples": 90761
+  },
+  "text": "and guarded their footsteps through wild adventures down to the fountain in the stable yard,",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__iPad__ch6__m1__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.1105,
+    "samples": 98652
+  },
+  "speech_duration": 4.110499858856201,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__iPad__ch6",
+  "speaker_id": "m1",
+  "env_id": "dds__office2__iPad__ch6",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 26
+}
\ No newline at end of file
diff --git a/by-dataset/dds/sample_028/record.json b/by-dataset/dds/sample_028/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..8be7a73744f652bc234e1cf255d45427080bdf2a
--- /dev/null
+++ b/by-dataset/dds/sample_028/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "dds__office2__Uber__ch1__m10__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.1416,
+    "samples": 99398
+  },
+  "environment_audio_duration": 4.141583442687988,
+  "environment_audio_text": "｡ｰAfter such a fall as this, I shall think nothing of tumbling down-stairs!",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "dds__office2__Uber__ch1__m10__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.9486,
+    "samples": 118766
+  },
+  "speaker_audio_duration": 4.948583126068115,
+  "speaker_audio_text": "Buck lived at a big house in the sun-kissed Santa Clara Valley.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "dds__office2__Uber__ch1__m10__spk_enhanced.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.9486,
+    "samples": 118766
+  },
+  "text": "but to her great disappointment it was empty:",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "dds__office2__Uber__ch1__m10__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.2673,
+    "samples": 78414
+  },
+  "speech_duration": 3.2672500610351562,
+  "language": "en",
+  "dataset": "dds",
+  "conversation_id": "office2__Uber__ch1",
+  "speaker_id": "m10",
+  "env_id": "dds__office2__Uber__ch1",
+  "text_source": "original",
+  "asr_token_count": 0,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 28
+}
\ No newline at end of file
diff --git a/by-dataset/summary.json b/by-dataset/summary.json
new file mode 100644
index 0000000000000000000000000000000000000000..d71aef7131a6efd15519c85e7aa785756a94978e
--- /dev/null
+++ b/by-dataset/summary.json
@@ -0,0 +1,33 @@
+{
+  "repo": "ChristianYang/Env-TTS-Clean",
+  "n_per_dataset": 30,
+  "saved_per_dataset": {
+    "aishell4": 30,
+    "alimeeting": 30,
+    "ami": 30,
+    "msdwild": 30,
+    "m3sd": 30,
+    "chime6": 30,
+    "dds": 30
+  },
+  "group_dataset_map": {
+    "data/group_00040": "aishell4",
+    "data/group_00058": "alimeeting",
+    "data/group_00000": "ami",
+    "data/group_00050": "msdwild",
+    "data/group_00001": "ami",
+    "data/group_00010": "m3sd",
+    "data/group_00020": "m3sd",
+    "data/group_00011": "m3sd",
+    "data/group_00029": "m3sd",
+    "data/group_00059": "chime6",
+    "data/group_00048": "m3sd",
+    "data/group_00012": "aishell4",
+    "data/group_00002": "ami",
+    "data/group_00030": "m3sd",
+    "data/group_00021": "m3sd",
+    "data/group_00038": "msdwild",
+    "data/group_00091": "dds"
+  },
+  "elapsed_sec": 611.88
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_000/record.json b/by-dataset/voxconverse/sample_000/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..6a08d24dc72312624494070312e9250c2ab38c1f
--- /dev/null
+++ b/by-dataset/voxconverse/sample_000/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__aggyz__spk05__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 10.7,
+    "samples": 256800
+  },
+  "environment_audio_duration": 10.699999809265137,
+  "environment_audio_text": "The difference between what I support and what you support, Vice President Biden, is that you require them to opt in, and I would not require them to opt in. They would automatically be enrolled. They wouldn't have to buy in.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__aggyz__spk05__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 9.46,
+    "samples": 227040
+  },
+  "speaker_audio_duration": 9.460000038146973,
+  "speaker_audio_text": "In the United States of America, we are spending twice as much per capita on healthcare as the Canadians or any other major country on Earth.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__aggyz__spk05__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 9.46,
+    "samples": 227040
+  },
+  "text": "Yeah, but Americans don't want to pay twice as much as other countries. What people want is cost-effective healthcare. Medicare for all will save the average American substantial sums of money on his or her.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__aggyz__spk05__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "aggyz",
+  "speaker_id": "spk05",
+  "env_id": "aggyz",
+  "text_source": "asr",
+  "asr_token_count": 41,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 0
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_001/record.json b/by-dataset/voxconverse/sample_001/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..65454d3d362c7e8f26b4a4b227c053f90b94a503
--- /dev/null
+++ b/by-dataset/voxconverse/sample_001/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__afjiv__spk04__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "A leader in the organisation, you understand what digitisation means. You take the time to read widely in the sector. There are a lot of really good books. Kevin Kelly, who started Wired magazine, has written a great, great book on on various technologies. I think understanding the",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__afjiv__spk04__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.28,
+    "samples": 102720
+  },
+  "speaker_audio_duration": 4.28000020980835,
+  "speaker_audio_text": "Because rather than just sending them forms, we won't bite you.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__afjiv__spk04__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.28,
+    "samples": 102720
+  },
+  "text": "The different processes and their software names, so.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__afjiv__spk04__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.48,
+    "samples": 83520
+  },
+  "speech_duration": 3.4800000190734863,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "afjiv",
+  "speaker_id": "spk04",
+  "env_id": "afjiv",
+  "text_source": "asr",
+  "asr_token_count": 10,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 1
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_006/record.json b/by-dataset/voxconverse/sample_006/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..f2c5d0524325139b3a6859940bc5cc14a295d6d6
--- /dev/null
+++ b/by-dataset/voxconverse/sample_006/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__aepyx__spk01__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.24,
+    "samples": 101760
+  },
+  "environment_audio_duration": 4.239999771118164,
+  "environment_audio_text": "Today in New Hampshire, launching a fresh attack at Trump for favoring big government.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__aepyx__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.44,
+    "samples": 82560
+  },
+  "speaker_audio_duration": 3.440000057220459,
+  "speaker_audio_text": "We have a movement going on, folks. We have a movement. This is not.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__aepyx__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.44,
+    "samples": 82560
+  },
+  "text": "With it being a Canadian citizen, he said, \"Oh, I didn't know that.\"",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__aepyx__spk01__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.29,
+    "samples": 102960
+  },
+  "speech_duration": 4.289999961853027,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "aepyx",
+  "speaker_id": "spk01",
+  "env_id": "aepyx",
+  "text_source": "asr",
+  "asr_token_count": 19,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 6
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_007/record.json b/by-dataset/voxconverse/sample_007/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..fbf5591f754cae1cd0bdbb23baaf05462677dee8
--- /dev/null
+++ b/by-dataset/voxconverse/sample_007/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__aisvi__spk00__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.12,
+    "samples": 74880
+  },
+  "environment_audio_duration": 3.119999885559082,
+  "environment_audio_text": "Let's continue to look to the courts.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__aisvi__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 11.2,
+    "samples": 268800
+  },
+  "speaker_audio_duration": 11.199999809265137,
+  "speaker_audio_text": "In New York, meanwhile, state lawmakers passed a bill that would give Congress access to the president's state tax returns if the IRS does not release them in full.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__aisvi__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 11.2,
+    "samples": 268800
+  },
+  "text": "Says tax returns must be given to Congress unless executive privilege is invoked. Before the House Financial Services Committee today, Treasury Secretary Stephen Mnuchin said the memo doesn't contradict the Treasury's.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__aisvi__spk00__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "aisvi",
+  "speaker_id": "spk00",
+  "env_id": "aisvi",
+  "text_source": "asr",
+  "asr_token_count": 39,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 7
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_008/record.json b/by-dataset/voxconverse/sample_008/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..14383cfa87c17422bac2e7ca0f2cd82638c5b546
--- /dev/null
+++ b/by-dataset/voxconverse/sample_008/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__aisvi__spk00__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.68,
+    "samples": 88320
+  },
+  "environment_audio_duration": 3.680000066757202,
+  "environment_audio_text": "And we believe that the president of the United States is engaged in a cover-up.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__aisvi__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 14.84,
+    "samples": 356160
+  },
+  "speaker_audio_duration": 14.84000015258789,
+  "speaker_audio_text": "And there was one area of cooperation between Congress and the executive branch. The House Intelligence Committee says it reached a deal with the Justice Department to turn over some underlying intelligence documents from the Mueller report.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__aisvi__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 14.84,
+    "samples": 356160
+  },
+  "text": "In New York, meanwhile, state lawmakers passed a bill that would give Congress access to the president's state tax returns if the IRS does not release them in full.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__aisvi__spk00__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 11.2,
+    "samples": 268800
+  },
+  "speech_duration": 11.199999809265137,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "aisvi",
+  "speaker_id": "spk00",
+  "env_id": "aisvi",
+  "text_source": "asr",
+  "asr_token_count": 33,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 8
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_009/record.json b/by-dataset/voxconverse/sample_009/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..29cf6dd8942df0b085b8e6318d3001e05a4adf69
--- /dev/null
+++ b/by-dataset/voxconverse/sample_009/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__ahnss__spk03__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 9.12,
+    "samples": 218880
+  },
+  "environment_audio_duration": 9.119999885559082,
+  "environment_audio_text": "And if I could make a quick point on my thoughts, just on Elizabeth Warren, I think it's really interesting that this week she's come out and identified as her target Facebook, and and really in this kind of.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__ahnss__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "But those two, and I hope they give them enough time, and I hope they devote enough time to foreign policy, and and don't ask the same stereotypical questions that some of the other moderators have fallen into. And of course, tomorrow night's debate also features billionaire Tom Steyer. Believe it.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__ahnss__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "And I think so because, and you know, Andrew Yang has tried to go for those big moments. Tom Steyer absolutely has to, because yes, Democrats know who he is because he's got a lot of money and he's a one or two issue kind of guy, particularly climate change and impeachment.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__ahnss__spk03__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "ahnss",
+  "speaker_id": "spk03",
+  "env_id": "ahnss",
+  "text_source": "asr",
+  "asr_token_count": 62,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 9
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_010/record.json b/by-dataset/voxconverse/sample_010/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..4dec72933aa1e2a14206c77aa0d6d218775cd73a
--- /dev/null
+++ b/by-dataset/voxconverse/sample_010/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__ahnss__spk03__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.0,
+    "samples": 120000
+  },
+  "environment_audio_duration": 5.0,
+  "environment_audio_text": "She was right, but her record did not reflect. At the time, she was always trying to be a progressive prosecutor, and that's never.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__ahnss__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "And I think so because, and you know, Andrew Yang has tried to go for those big moments. Tom Steyer absolutely has to, because yes, Democrats know who he is because he's got a lot of money and he's a one or two issue kind of guy, particularly climate change and impeachment.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__ahnss__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "Bernie's not going after Biden the way he's going after Liz. That he knows he has to make a big move. She is the biggest threat. He still got a lot of money. He doesn't have to get out of the race right now. He's not like some of these, like Eric Swalwell.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__ahnss__spk03__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.72,
+    "samples": 305280
+  },
+  "speech_duration": 12.720000267028809,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "ahnss",
+  "speaker_id": "spk03",
+  "env_id": "ahnss",
+  "text_source": "asr",
+  "asr_token_count": 66,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 10
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_012/record.json b/by-dataset/voxconverse/sample_012/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..8edaf56bba1a59db44accb9ce31ca6fc79eb2e3c
--- /dev/null
+++ b/by-dataset/voxconverse/sample_012/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__akthc__spk00__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "Almost over. The first televised face-off between Hillary Clinton and Donald Trump will get underway in less than four hours from now. It's expected to be the most watched presidential debate in history and could even rival ratings seen during the Super Bowl.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__akthc__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 10.16,
+    "samples": 243840
+  },
+  "speaker_audio_duration": 10.15999984741211,
+  "speaker_audio_text": "Democratic presidential candidate Hillary Clinton and her Republican rival Donald Trump are gearing up for their first debate on Monday night in one of the century's most anticipated TV events.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__akthc__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 10.16,
+    "samples": 243840
+  },
+  "text": "Moderated by NBC News anchor Lester Holt, it's a 90-minute session starting at 9 p.m. at New York's Hofstra University.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__akthc__spk00__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.24,
+    "samples": 173760
+  },
+  "speech_duration": 7.239999771118164,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "akthc",
+  "speaker_id": "spk00",
+  "env_id": "akthc",
+  "text_source": "asr",
+  "asr_token_count": 32,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 12
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_013/record.json b/by-dataset/voxconverse/sample_013/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..e0008d341a0667764652e9b928e6bb898da8013a
--- /dev/null
+++ b/by-dataset/voxconverse/sample_013/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__akthc__spk00__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "Almost over. The first televised face-off between Hillary Clinton and Donald Trump will get underway in less than four hours from now. It's expected to be the most watched presidential debate in history and could even rival ratings seen during the Super Bowl.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__akthc__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "Just six weeks ahead of election day, the polls show the two candidates neck and neck. A poll release on Sunday by ABC News and the Washington Post showed voters are split, with 46 percent for Clinton and 44 percent for Trump. Well within the polls'",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__akthc__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "Clinton has three decades of experience in politics and a long history of high-profile debates.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__akthc__spk00__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.52,
+    "samples": 132480
+  },
+  "speech_duration": 5.519999980926514,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "akthc",
+  "speaker_id": "spk00",
+  "env_id": "akthc",
+  "text_source": "asr",
+  "asr_token_count": 17,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 13
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_014/record.json b/by-dataset/voxconverse/sample_014/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..201cf48739519c3a83a094bf5f70bbd96de17ad6
--- /dev/null
+++ b/by-dataset/voxconverse/sample_014/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__azisu__spk00__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.88,
+    "samples": 189120
+  },
+  "environment_audio_duration": 7.880000114440918,
+  "environment_audio_text": "Yeah, well, you can look at these things, and you can look at the climate realities, and even this NCA report says that the the trend on tornadoes isn't certain. The IPCC report.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__azisu__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "Well, you get the message out that this is serious business, and you know if you live in Oklahoma, where tornadoes have wiped your town out a couple times, if you and you chose Alaska, which is remote generally, but when you start, you remember Hurricane Sandy, the bottom.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__azisu__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "But you're talking about millions of tons of road that have to be lifted, and that energy has to come from somewhere, and that's just the start of things.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__azisu__spk00__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 7.44,
+    "samples": 178560
+  },
+  "speech_duration": 7.440000057220459,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "azisu",
+  "speaker_id": "spk00",
+  "env_id": "azisu",
+  "text_source": "asr",
+  "asr_token_count": 34,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 14
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_015/record.json b/by-dataset/voxconverse/sample_015/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..95c5c02cff5856fd4d0023d5b51ce5e960b653c9
--- /dev/null
+++ b/by-dataset/voxconverse/sample_015/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__auzru__spk01__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "You see, you need 1,991 delegates to secure the Democratic nomination. Look what happened with Joe Biden tonight. He just jumped ahead of Pete Buttigieg, and he—that's with 14 delegates. We still have 40 to allocate from South Carolina. You've got Bernie.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__auzru__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.99,
+    "samples": 143760
+  },
+  "speaker_audio_duration": 5.989999771118164,
+  "speaker_audio_text": "Is it really? It's really doubtful at this point that it'll happen before Super Tuesday. But you never know. We'll see how it.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__auzru__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.99,
+    "samples": 143760
+  },
+  "text": "And even South Carolina, the question always was.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__auzru__spk01__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.2,
+    "samples": 76800
+  },
+  "speech_duration": 3.200000047683716,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "auzru",
+  "speaker_id": "spk01",
+  "env_id": "auzru",
+  "text_source": "asr",
+  "asr_token_count": 10,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 15
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_017/record.json b/by-dataset/voxconverse/sample_017/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..7389bbab770577af0925386e43eadfa3549de7a4
--- /dev/null
+++ b/by-dataset/voxconverse/sample_017/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__auzru__spk02__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.45,
+    "samples": 82800
+  },
+  "environment_audio_duration": 3.450000047683716,
+  "environment_audio_text": "So let's get over to Wolf, who has some new information about delegates.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__auzru__spk02__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "And that's the hard part for Joe Biden, in the sense that Bloomberg is definitely not getting up before Super Tuesday. He's not on the ballot until Super Tuesday. His whole strategy is built around Super Tuesday. So what Biden gets a win tonight, the question is how does it translate in the next 72 hours? Because he so desperately would need to follow up on Tuesday with more.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__auzru__spk02__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "Electorate in 2016 in Alabama was African American. It was 32% in North Carolina, 54% in Alabama, 19% in Texas. Is this a springboard? Because Joe Biden, even if he had some money, even if he raised a boatload of money tonight, can't get ads on TV fast enough in these super two.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__auzru__spk02__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "auzru",
+  "speaker_id": "spk02",
+  "env_id": "auzru",
+  "text_source": "asr",
+  "asr_token_count": 80,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 17
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_019/record.json b/by-dataset/voxconverse/sample_019/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..68907dc61ae8cb81688bf656d2d73d799c77e994
--- /dev/null
+++ b/by-dataset/voxconverse/sample_019/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk00__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "To recover from their childhood, good advice. But if we're going to do the same thing with teenagers, we're going to need a resilient strategy to help us figure out which one to choose. I'm joined by clinical psychologist and family therapist Andrew Fuller, teacher, nurse.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.92,
+    "samples": 214080
+  },
+  "speaker_audio_duration": 8.920000076293945,
+  "speaker_audio_text": "But in comparison to some other resilience programs, this one is absolutely easy. All you need is a three by five space, some lighter fluid, and a rake.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bdopb__spk00__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 8.92,
+    "samples": 214080
+  },
+  "text": "You use the term fire pit, and people immediately think of.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bdopb__spk00__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.28,
+    "samples": 78720
+  },
+  "speech_duration": 3.2799999713897705,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bdopb",
+  "speaker_id": "spk00",
+  "env_id": "bdopb",
+  "text_source": "asr",
+  "asr_token_count": 13,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 19
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_021/record.json b/by-dataset/voxconverse/sample_021/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..3078b89aa0414b57f3e4f712e8b42472749c9e9d
--- /dev/null
+++ b/by-dataset/voxconverse/sample_021/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bauzd__spk03__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.08,
+    "samples": 73920
+  },
+  "environment_audio_duration": 3.0799999237060547,
+  "environment_audio_text": "He's not gonna stay on national TV and say black people are prone to criminality.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bauzd__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "We have the unconscious bias, and it is proven over and over again. And no, it is not just because ah, you know, the blacks are prone to criminality. No, you're targeting blacks more, and it's proven out by the data. But you don't want to see the data. You don't want.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bauzd__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "When they removed all the factors, is it a violent city? Is there more crime in that city? Was the cop in imminent danger? You know what? The one factor that led most conclusively to whether somebody was going to get shot by a police officer.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bauzd__spk03__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 12.64,
+    "samples": 303360
+  },
+  "speech_duration": 12.640000343322754,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bauzd",
+  "speaker_id": "spk03",
+  "env_id": "bauzd",
+  "text_source": "asr",
+  "asr_token_count": 53,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 21
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_022/record.json b/by-dataset/voxconverse/sample_022/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..becadc9290a90619bea9f4d74aca2b03b834ea76
--- /dev/null
+++ b/by-dataset/voxconverse/sample_022/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk03__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.84,
+    "samples": 140160
+  },
+  "environment_audio_duration": 5.840000152587891,
+  "environment_audio_text": "Do you think that's something particularly that it happens in Australia, or would it happen in other countries? Do you think?",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.72,
+    "samples": 329280
+  },
+  "speaker_audio_duration": 13.720000267028809,
+  "speaker_audio_text": "Yeah, that brings that on. Am I going to get the marks that I want? Am I going to get into the course that I want or the job that I want? So it's incredibly vulnerable time. So much so that we really have to rethink our model around well-being, particularly at year 11 and year 12.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bdopb__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.72,
+    "samples": 329280
+  },
+  "text": "Oh, when we've surveyed over 22,000 young Australians in terms of this, and you do see incremental shifts in terms of those eight key areas that I mentioned earlier, as schools embark upon initiatives. But beyond the the stats.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bdopb__spk03__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bdopb",
+  "speaker_id": "spk03",
+  "env_id": "bdopb",
+  "text_source": "asr",
+  "asr_token_count": 51,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 22
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_023/record.json b/by-dataset/voxconverse/sample_023/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..3eced776d29c03ddd07a820f953e44157fbaf3d7
--- /dev/null
+++ b/by-dataset/voxconverse/sample_023/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk03__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 5.84,
+    "samples": 140160
+  },
+  "environment_audio_duration": 5.840000152587891,
+  "environment_audio_text": "Do you think that's something particularly that it happens in Australia, or would it happen in other countries? Do you think?",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "Critically, is helping parents as well as students understand how to succeed at school is important. That that journey often begins around the middle of year ten. is often the time when schools really need to reconnect with the parents that they've lost along the way. But the other thing.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bdopb__spk03__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "Is that forgiveness creeps into a school? Forgiveness is a critical key in all of this because I suppose the results of those 22,000 surveys that we've done tell us that young Australians are quite good at forming relationships. What they're not very",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bdopb__spk03__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bdopb",
+  "speaker_id": "spk03",
+  "env_id": "bdopb",
+  "text_source": "asr",
+  "asr_token_count": 55,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 23
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_024/record.json b/by-dataset/voxconverse/sample_024/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..e1e8d06c00a12fef56ffa69671d290294bea7f9e
--- /dev/null
+++ b/by-dataset/voxconverse/sample_024/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk01__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "For all schools, and the way to achieve that can vary. It can be a kind of continuous program, as Martin said, or other schools sort of will do it in very different ways. And I think that's fine. There's many, many ways to achieve the same outcomes. Some schools will decide we're busy.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.68,
+    "samples": 88320
+  },
+  "speaker_audio_duration": 3.680000066757202,
+  "speaker_audio_text": "Another one is attendance. Attendance from staff and students.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bdopb__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 3.68,
+    "samples": 88320
+  },
+  "text": "It has, yeah. The challenge we had is it can't look like tokenism, so you can't just fix something by implementing a program. It has to be part of the culture of the school too. So we did a lot of work, not just the program, but also in terms of changing the language staff.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bdopb__spk01__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bdopb",
+  "speaker_id": "spk01",
+  "env_id": "bdopb",
+  "text_source": "asr",
+  "asr_token_count": 66,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 24
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_025/record.json b/by-dataset/voxconverse/sample_025/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..82824c12b2f77bc2432319624c91b38b44de596a
--- /dev/null
+++ b/by-dataset/voxconverse/sample_025/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk01__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.72,
+    "samples": 329280
+  },
+  "environment_audio_duration": 13.720000267028809,
+  "environment_audio_text": "Yeah, that brings that on. Am I going to get the marks that I want? Am I going to get into the course that I want or the job that I want? So it's incredibly vulnerable time. So much so that we really have to rethink our model around well-being, particularly at year 11 and year 12.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "Well, for the for the information for our life skills subject, the resiliency stuff, I actually swapped what we're doing with our school with someone that I met at a PD from another school, and we just exchanged our programs and and talked about what worked and what didn't work. So kind of.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bdopb__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "But then we've got the unofficial, which is where it's become part of the culture of the school.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bdopb__spk01__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 4.44,
+    "samples": 106560
+  },
+  "speech_duration": 4.440000057220459,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bdopb",
+  "speaker_id": "spk01",
+  "env_id": "bdopb",
+  "text_source": "asr",
+  "asr_token_count": 22,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 25
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_026/record.json b/by-dataset/voxconverse/sample_026/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..4d14628f96102918d03507160e56f632b5369c85
--- /dev/null
+++ b/by-dataset/voxconverse/sample_026/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk01__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.36,
+    "samples": 320640
+  },
+  "environment_audio_duration": 13.359999656677246,
+  "environment_audio_text": "So it's recognising positive behaviour, and for each time you are recognised for positive behaviour, you get a merit card. Or we actually use an online modelling system where we can just tick and flick a box.",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__bdopb__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.2,
+    "samples": 316800
+  },
+  "speaker_audio_duration": 13.199999809265137,
+  "speaker_audio_text": "Andrew brought up just reminded me that it's it's really important to review too, and to keep reviewing. And you can't just implement a program and go problem solved. We're done now. You you need to keep checking back and reviewing how it's going, it and and tweaking it.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__bdopb__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 13.2,
+    "samples": 316800
+  },
+  "text": "Well, for the for the information for our life skills subject, the resiliency stuff, I actually swapped what we're doing with our school with someone that I met at a PD from another school, and we just exchanged our programs and and talked about what worked and what didn't work. So kind of.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__bdopb__spk01__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "bdopb",
+  "speaker_id": "spk01",
+  "env_id": "bdopb",
+  "text_source": "asr",
+  "asr_token_count": 63,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 26
+}
\ No newline at end of file
diff --git a/by-dataset/voxconverse/sample_028/record.json b/by-dataset/voxconverse/sample_028/record.json
new file mode 100644
index 0000000000000000000000000000000000000000..6b9f5658b066407b16b48d4b43067560d2d750bf
--- /dev/null
+++ b/by-dataset/voxconverse/sample_028/record.json
@@ -0,0 +1,51 @@
+{
+  "_files": {
+    "environment_audio_source": "environment_audio_source.wav",
+    "speaker_audio_source": "speaker_audio_source.wav",
+    "speaker_audio_source_enhanced": "speaker_audio_source_enhanced.wav",
+    "speech": "speech.wav"
+  },
+  "environment_audio_source": {
+    "wav": "environment_audio_source.wav",
+    "source_path": "voxconverse__blwmj__spk01__env.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "environment_audio_duration": 15.0,
+  "environment_audio_text": "2020 senior legal advisor, constitutional law attorney joins us. So Jenna, the complaint focuses on a CNN opinion story by a former Federal Election Commission general counsel, Larry Noble. He wrote a piece titled, \"Quote: Soliciting dirt on your opponents from a foreign government.\"",
+  "speaker_audio_source": {
+    "wav": "speaker_audio_source.wav",
+    "source_path": "voxconverse__blwmj__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speaker_audio_duration": 15.0,
+  "speaker_audio_text": "Well, we're suing CNN as well as the New York Times and Washington Post for false and defamatory statements. And CNN, in particular, claimed that the campaign had assessed the potential risks and benefits of again soliciting Russia's help in the 2020 election.",
+  "speaker_audio_source_enhanced": {
+    "wav": "speaker_audio_source_enhanced.wav",
+    "source_path": "voxconverse__blwmj__spk01__spk.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "text": "This was published in the context of an opinion piece. The court has also had a long precedent of case law that just because you couch a statement as opinion doesn't mean that you can assert and make false statements of fact. The court will look at each individual statement and determine.",
+  "speech": {
+    "wav": "speech.wav",
+    "source_path": "voxconverse__blwmj__spk01__speech.flac",
+    "sample_rate_hz": 24000,
+    "duration_sec": 15.0,
+    "samples": 360000
+  },
+  "speech_duration": 15.0,
+  "language": "en",
+  "dataset": "voxconverse",
+  "conversation_id": "blwmj",
+  "speaker_id": "spk01",
+  "env_id": "blwmj",
+  "text_source": "asr",
+  "asr_token_count": 56,
+  "asr_mean_logprob": 0.0,
+  "_sample_index": 28
+}
\ No newline at end of file
diff --git a/env-tts-clean/summary.json b/env-tts-clean/summary.json
new file mode 100644
index 0000000000000000000000000000000000000000..355f08899dff4381f30a803bdfaa0bda6be247cd
--- /dev/null
+++ b/env-tts-clean/summary.json
@@ -0,0 +1,67 @@
+{
+  "repo": "ChristianYang/Env-TTS-Clean",
+  "n_requested": 50,
+  "n_saved": 50,
+  "seed": 42,
+  "shuffle_buffer": 10000,
+  "samples": [
+    "sample_000",
+    "sample_001",
+    "sample_002",
+    "sample_003",
+    "sample_004",
+    "sample_005",
+    "sample_006",
+    "sample_007",
+    "sample_008",
+    "sample_009",
+    "sample_010",
+    "sample_011",
+    "sample_012",
+    "sample_013",
+    "sample_014",
+    "sample_015",
+    "sample_016",
+    "sample_017",
+    "sample_018",
+    "sample_019",
+    "sample_020",
+    "sample_021",
+    "sample_022",
+    "sample_023",
+    "sample_024",
+    "sample_025",
+    "sample_026",
+    "sample_027",
+    "sample_028",
+    "sample_029",
+    "sample_030",
+    "sample_031",
+    "sample_032",
+    "sample_033",
+    "sample_034",
+    "sample_035",
+    "sample_036",
+    "sample_037",
+    "sample_038",
+    "sample_039",
+    "sample_040",
+    "sample_041",
+    "sample_042",
+    "sample_043",
+    "sample_044",
+    "sample_045",
+    "sample_046",
+    "sample_047",
+    "sample_048",
+    "sample_049"
+  ],
+  "dataset_distribution": {
+    "ami": 5,
+    "m3sd": 32,
+    "aishell4": 5,
+    "alimeeting": 6,
+    "msdwild": 2
+  },
+  "audio_decode_errors": []
+}
\ No newline at end of file