whisper-tiny.en-atcosim
This model is a fine-tuned version of openai/whisper-tiny.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0737
- Wer: 2.7135
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4068 | 8.33 | 500 | 0.0639 | 2.4958 |
| 0.0021 | 16.67 | 1000 | 0.0638 | 2.3245 |
| 0.0006 | 25.0 | 1500 | 0.0665 | 2.3338 |
| 0.0003 | 33.33 | 2000 | 0.0680 | 2.2736 |
| 0.0002 | 41.67 | 2500 | 0.0693 | 2.3893 |
| 0.0001 | 50.0 | 3000 | 0.0703 | 2.4634 |
| 0.0001 | 58.33 | 3500 | 0.0713 | 2.4449 |
| 0.0001 | 66.67 | 4000 | 0.0720 | 2.4542 |
| 0.0001 | 75.0 | 4500 | 0.0727 | 2.4588 |
| 0.0001 | 83.33 | 5000 | 0.0732 | 2.6394 |
| 0.0001 | 91.67 | 5500 | 0.0736 | 2.7783 |
| 0.0001 | 100.0 | 6000 | 0.0737 | 2.7135 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2
- Datasets 2.15.0
- Tokenizers 0.15.0