openai/whisper-large-v2
This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2381
- Wer: 11.1244
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.3639 | 0.12 | 500 | 0.2512 | 12.9597 |
| 0.1931 | 0.25 | 1000 | 0.2123 | 12.1414 |
| 0.329 | 1.08 | 1500 | 0.2064 | 11.5818 |
| 0.097 | 1.21 | 2000 | 0.2050 | 10.9775 |
| 0.0522 | 2.04 | 2500 | 0.2258 | 10.4390 |
| 0.1026 | 2.17 | 3000 | 0.2201 | 11.7017 |
| 0.0448 | 3.0 | 3500 | 0.2287 | 10.3873 |
| 0.0455 | 3.13 | 4000 | 0.2381 | 11.1244 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2