Whisper small ID - Augmented
This model is a fine-tuned version of openai/whisper-small on the YT dataset. It achieves the following results on the evaluation set:
- Loss: 0.2957
- Wer: 24.0697
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: 6
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 96
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.5857 | 0.52 | 500 | 0.4354 | 39.9358 |
| 0.385 | 1.03 | 1000 | 0.3511 | 31.3213 |
| 0.2967 | 1.55 | 1500 | 0.3075 | 25.8541 |
| 0.1548 | 2.06 | 2000 | 0.2957 | 24.0697 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 2.0.0+cu117
- Datasets 2.10.1
- Tokenizers 0.13.2