Hugging Face's logo

ltuncay
/
BEST-RQ-2.1-base

Feature Extraction
Transformers
Safetensors
audio_embeddings
audio
custom_code
self-supervised-learning
audio-embeddings
best-rq-2
audioset
Model card Files
xet
 Community
ltuncay commited on
Commit
eddde3f
·
verified ·
1 Parent(s): a7d8813

Link comparison-table entries to published Hugging Face models

Browse files
Files changed (2) hide show
  1. README.md +6 -6
  2. export_manifest.json +1 -1
README.md CHANGED
@@ -29,16 +29,16 @@ BEST-RQ (Conformer), BEST-RQ (ViT),
29
  and all BEST-RQ-2 variants reported below are trained on the **same AudioSet
30
  split for 200,000 steps**. The pretrained baselines are shown for comparison.
31
 
32
- | Model | Speech | Music | Environment | Global Mean | Mean of Means | Training recipe |
33
  | --- | ---: | ---: | ---: | ---: | ---: | --- |
34
  | data2vec | 50.62 | 23.24 | 15.41 | 37.83 | 29.76 | Pretrained baseline |
35
  | wav2vec 2.0 | 41.79 | 34.94 | 29.52 | 37.84 | 35.42 | Pretrained baseline |
36
  | Whisper | 49.19 | 38.67 | 28.61 | 42.75 | 38.82 | Pretrained baseline |
37
- | BEST-RQ (Conformer) | 40.43 | 35.58 | 30.81 | 37.43 | 35.60 | Separate codebase (recipe unavailable) |
38
- | BEST-RQ (ViT) | 32.87 | 41.62 | 34.50 | 34.88 | 36.33 | [best_rq/audioset/default.yaml](https://github.com/LudovicTuncay/audio-embeddings/blob/bb88bf790b1dcf8251c6b38e7a4766534adf33d3/configs/experiment/best_rq/audioset/default.yaml) |
39
- | BEST-RQ-2 (Interspeech 2026) | 38.49 | 54.40 | 46.39 | 43.21 | 46.43 | [best_rq_2/default.yaml](https://github.com/LudovicTuncay/audio-embeddings/blob/bb88bf790b1dcf8251c6b38e7a4766534adf33d3/configs/experiment/best_rq_2/default.yaml) |
40
- | BEST-RQ-2.1 | 52.60 | 62.23 | 53.38 | 54.59 | 56.07 | [best_rq_2_1/masking/80.yaml](https://github.com/LudovicTuncay/audio-embeddings/blob/bb88bf790b1dcf8251c6b38e7a4766534adf33d3/configs/experiment/best_rq_2_1/masking/80.yaml) |
41
- | BEST-RQ-2.2 | **53.78** | **63.90** | **55.71** | **56.11** | **57.80** | [best_rq_2_2/masking/80.yaml](https://github.com/LudovicTuncay/audio-embeddings/blob/bb88bf790b1dcf8251c6b38e7a4766534adf33d3/configs/experiment/best_rq_2_2/masking/80.yaml) |
42
 
43
  Global Mean averages all benchmark task scores. Mean of Means gives equal
44
  weight to the Speech, Music, and Environment category means.
 
29
  and all BEST-RQ-2 variants reported below are trained on the **same AudioSet
30
  split for 200,000 steps**. The pretrained baselines are shown for comparison.
31
 
32
+ | Model | Speech | Music | Environment | Global Mean | Mean of Means | Hugging Face model |
33
  | --- | ---: | ---: | ---: | ---: | ---: | --- |
34
  | data2vec | 50.62 | 23.24 | 15.41 | 37.83 | 29.76 | Pretrained baseline |
35
  | wav2vec 2.0 | 41.79 | 34.94 | 29.52 | 37.84 | 35.42 | Pretrained baseline |
36
  | Whisper | 49.19 | 38.67 | 28.61 | 42.75 | 38.82 | Pretrained baseline |
37
+ | BEST-RQ (Conformer) | 40.43 | 35.58 | 30.81 | 37.43 | 35.60 | Separate codebase |
38
+ | BEST-RQ (ViT) | 32.87 | 41.62 | 34.50 | 34.88 | 36.33 | [BEST-RQ-ViT](https://huggingface.co/ltuncay/BEST-RQ-ViT) |
39
+ | BEST-RQ-2 (Interspeech 2026) | 38.49 | 54.40 | 46.39 | 43.21 | 46.43 | [BEST-RQ-2](https://huggingface.co/ltuncay/BEST-RQ-2) |
40
+ | BEST-RQ-2.1 | 52.60 | 62.23 | 53.38 | 54.59 | 56.07 | [BEST-RQ-2.1-base](https://huggingface.co/ltuncay/BEST-RQ-2.1-base) |
41
+ | BEST-RQ-2.2 | **53.78** | **63.90** | **55.71** | **56.11** | **57.80** | [BEST-RQ-2.2-base](https://huggingface.co/ltuncay/BEST-RQ-2.2-base) |
42
 
43
  Global Mean averages all benchmark task scores. Mean of Means gives equal
44
  weight to the Speech, Music, and Environment category means.
export_manifest.json CHANGED
@@ -31,7 +31,7 @@
31
  },
32
  "files": {
33
  "CODE_LICENSE": "8fe9e8b749cd4abedabcb3100df445db899e72192394d14cc2dbf24a40811af6",
34
- "README.md": "194db8166e421ac7fa8709099f16fa7a9ac476d21d01dd03527aca66ca3a78fc",
35
  "adapters.py": "ca0f26826763c0e48b7508da732242bb83adfeb4814e389ed3fde93ad648c728",
36
  "config.json": "88be4a7618663564525588db633cf7461f051e437913187070b854ae8367ac07",
37
  "configuration_audio.py": "59f3a0b8db0df5af85e677ac33a4431595e9b2eff6d34c1e6a1778dd75a4568d",
 
31
  },
32
  "files": {
33
  "CODE_LICENSE": "8fe9e8b749cd4abedabcb3100df445db899e72192394d14cc2dbf24a40811af6",
34
+ "README.md": "b3ec1944ff3384c44d5d240f693915908052528e006478e9bd5d81631c0a1b3a",
35
  "adapters.py": "ca0f26826763c0e48b7508da732242bb83adfeb4814e389ed3fde93ad648c728",
36
  "config.json": "88be4a7618663564525588db633cf7461f051e437913187070b854ae8367ac07",
37
  "configuration_audio.py": "59f3a0b8db0df5af85e677ac33a4431595e9b2eff6d34c1e6a1778dd75a4568d",