Record successful public inference with custom audio references

verification.json

@@ -1,64 +1,61 @@
 {
   "passed": true,
   "repo_id": "penguinfish1688/duplexdataengine",
-  "revision": "8d263feb33ec6c61ac9aa80241f18479924d62a0",
+  "revision": "ac37044a08f9b5ed6f6ce97b8153abc811c12e51",
   "anonymous_download": true,
   "isolated_environment": true,
   "research_checkout_required": false,
+  "model_version": "1.1.0",
   "gpu": "NVIDIA RTX PRO 6000 Blackwell Server Edition",
+  "custom_reference_audio": true,
   "torch": "2.8.0+cu128",
   "examples": [
     {
       "id": "sample_1",
+      "voice_inputs": [
+        "audio_file",
+        "preset"
+      ],
       "eos": [
         true,
         true
       ],
       "frames": [
-        582,
-        595
-      ],
-      "audio_seconds": 47.6,
-      "wall_seconds": 50.687844783999026,
-      "sha256": "a0d1689519dfee0ed034e9b0d5d6126ed6e6cc3d258da3d0be62793e57385085",
-      "correlation_with_featured_audio": [
-        0.9999125589163387,
-        0.9996439714705407
+        654,
+        659
       ],
-      "label": "Sample 1"
+      "audio_seconds": 52.72,
+      "wall_seconds": 57.37205607211217,
+      "sha256": "aa4d4afd52d9e3722ea839c8f71daf018963dd1ea6bccfb29a748a5535a46473"
     },
     {
       "id": "sample_2",
+      "voice_inputs": [
+        "audio_file",
+        "audio_file"
+      ],
       "eos": [
         true,
         true
       ],
       "frames": [
-        680,
-        660
+        662,
+        653
       ],
-      "audio_seconds": 54.4,
-      "wall_seconds": 57.355075997067615,
-      "sha256": "e616a52ff05ed2287a00d435d8430f3f0b214c7c4017ca6bed68d8725283f9d5",
-      "correlation_with_featured_audio": [
-        0.9999395525329423,
-        0.9998978182617578
-      ],
-      "label": "Sample 2"
+      "audio_seconds": 52.96,
+      "wall_seconds": 56.35323277302086,
+      "sha256": "16648457873745f96f1136a819386cbe6b323fea4f24623aca409e24be1b3d97"
     }
   ],
-  "peak_gpu_memory_gb": 11.553627648,
-  "seconds": 135.95635328697972,
+  "peak_gpu_memory_gb": 11.571285504,
+  "seconds": 115.65257979906164,
   "listening_performed": false,
-  "checked_at_utc": "2026-09-27T02:37:07.612676+00:00",
-  "checks": [
-    "Anonymous complete download",
-    "Isolated pip installation from pinned public dependencies",
-    "No research package installed",
-    "Both full featured XML inputs",
-    "Both lanes emit EOS",
-    "Finite normalized 24 kHz stereo waveform",
-    "Shared clock preserved; waveform correlation above 0.99 with each featured lane"
-  ],
-  "model_version": "1.0.0"
+  "checked_at_utc": "2026-09-27T03:05:35.570379+00:00",
+  "cpu_checks": {
+    "focused_tests_passed": 9,
+    "real_encoder_cpu_passed": true,
+    "reference_embedding_dimension": 2048,
+    "reference_cosine_similarity": 0.9999986290931702,
+    "trained_speaker_encoder_reused": true
+  }
 }