Downloads · 30 days
8
9% of all-time downloads
wav2vec2-bn-checkpoints
pawmeow/wav2vec2-bn-checkpoints
wav2vec2-bn-checkpoints is a automatic speech recognition model from pawmeow. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
90
Public
Parameters
316M
24 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- wav2vec2
- Access
- Public
- Created
- Nov 11, 2025
- Updated
- Nov 14, 2025
- SHA
- eba5f128
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
wav2vec2-bn-checkpoints
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.9099
- Norm Lev Sim Mean: 0.0187
- Wer: 1.0
- Cer: 0.9812
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Cer | Decode Mode | Validation Loss | Norm Lev Sim Mean | Wer |
|---|---|---|---|---|---|---|---|
| 20.1827 | 1.0 | 95 | 0.9775 | greedy | 26.9940 | 0.0230 | 1.0 |
| 18.5724 | 2.0 | 190 | 1.0594 | greedy | 25.8965 | 0.0261 | 1.0 |
| 18.9031 | 3.0 | 285 | 0.9740 | greedy | 23.9979 | 0.0265 | 1.0 |
| 14.6475 | 4.0 | 380 | 0.9683 | greedy | 21.1389 | 0.0323 | 1.0 |
| 14.3493 | 4.9973 | 470 | 17.5431 | 0.0321 | 1.0 | 0.9682 | |
| 11.6066 | 6.0 | 565 | 12.5254 | 0.0280 | 1.0 | 0.9725 | |
| 7.5114 | 7.0 | 660 | 8.1257 | 0.0254 | 1.0 | 0.9749 | |
| 5.8601 | 8.0 | 755 | 5.3355 | 0.0219 | 1.0 | 0.9780 | |
| 4.3301 | 9.0 | 850 | 4.2174 | 0.0187 | 1.0 | 0.9812 | |
| 4.1288 | 10.0 | 945 | 3.9223 | 0.0191 | 1.0 | 0.9808 | |
| 3.7416 | 11.0 | 1040 | 3.9099 | 0.0188 | 1.0 | 0.9812 | |
| 4.0712 | 12.0 | 1135 | 3.9099 | 0.0181 | 1.0 | 0.9818 | |
| 3.7301 | 13.0 | 1230 | 3.9099 | 0.0185 | 1.0 | 0.9814 | |
| 4.0605 | 14.0 | 1325 | 3.9098 | 0.0187 | 1.0 | 0.9812 | |
| 4.1702 | 14.9019 | 1410 | 3.9099 | 0.0187 | 1.0 | 0.9812 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- License
- apache-2.0
- Created
- Nov 11, 2025
- Updated
- Nov 14, 2025