openai/whisper-small
This model is a fine-tuned version of openai/whisper-small on the pphuc25/VietMed-split-8-2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.8268
- Wer: 39.2934
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.5567 | 1.0 | 569 | 0.5442 | 110.6681 |
| 0.3927 | 2.0 | 1138 | 0.5414 | 79.2641 |
| 0.2763 | 3.0 | 1707 | 0.5484 | 81.6145 |
| 0.1841 | 4.0 | 2276 | 0.5896 | 70.0897 |
| 0.1059 | 5.0 | 2845 | 0.6430 | 59.4838 |
| 0.0807 | 6.0 | 3414 | 0.6636 | 63.3132 |
| 0.0459 | 7.0 | 3983 | 0.7000 | 71.0086 |
| 0.0249 | 8.0 | 4552 | 0.7402 | 59.5643 |
| 0.0188 | 9.0 | 5121 | 0.7577 | 70.7926 |
| 0.0182 | 10.0 | 5690 | 0.7692 | 52.9599 |
| 0.0049 | 11.0 | 6259 | 0.7727 | 77.1664 |
| 0.0054 | 12.0 | 6828 | 0.7771 | 41.3619 |
| 0.0019 | 13.0 | 7397 | 0.7908 | 49.6650 |
| 0.0023 | 14.0 | 7966 | 0.8098 | 34.6403 |
| 0.001 | 15.0 | 8535 | 0.8050 | 41.5669 |
| 0.0005 | 16.0 | 9104 | 0.8027 | 35.4970 |
| 0.0004 | 17.0 | 9673 | 0.8123 | 39.3520 |
| 0.0007 | 18.0 | 10242 | 0.8224 | 43.5914 |
| 0.0002 | 19.0 | 10811 | 0.8250 | 39.2532 |
| 0.0001 | 20.0 | 11380 | 0.8268 | 39.2934 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 0