openai/whisper-tiny
This model is a fine-tuned version of openai/whisper-tiny on the common_voice_22_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6814
- Wer: 20.4561
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: 3.75e-05
- train_batch_size: 256
- eval_batch_size: 128
- seed: 42
- 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: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 100000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0003 | 43.1034 | 5000 | 0.5045 | 21.9783 |
| 0.0 | 86.2069 | 10000 | 0.6353 | 22.5083 |
| 0.0 | 129.3103 | 15000 | 0.5335 | 20.2152 |
| 0.0 | 172.4138 | 20000 | 0.6182 | 20.4882 |
| 0.0 | 215.5172 | 25000 | 0.7206 | 20.8466 |
| 0.0 | 258.6207 | 30000 | 0.8050 | 21.2193 |
| 0.0 | 301.7241 | 35000 | 0.6262 | 20.5338 |
| 0.0 | 344.8276 | 40000 | 0.6814 | 20.4561 |
Framework versions
- Transformers 4.52.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1