wav2vec2-base-scl-common-voice
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.7258
- Wer: 0.8720
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: 0.0001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 7.6627 | 4.8077 | 500 | 3.1642 | 1.0 |
| 1.9566 | 9.6154 | 1000 | 1.3902 | 0.9209 |
| 0.6485 | 14.4231 | 1500 | 1.4672 | 0.8966 |
| 0.4033 | 19.2308 | 2000 | 1.5901 | 0.9092 |
| 0.2760 | 24.0385 | 2500 | 1.6703 | 0.8878 |
| 0.2079 | 28.8462 | 3000 | 1.7262 | 0.8703 |
| 0.2079 | 30.0 | 3120 | 1.7258 | 0.8720 |
Framework versions
- Transformers 5.12.0
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2