metadata
language:
- en
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: openai/whisper-tiny
results: []
openai/whisper-tiny
This model is a fine-tuned version of openai/whisper-tiny on the pphuc25/EngMed dataset. It achieves the following results on the evaluation set:
- Loss: 3.4853
- Wer: 66.7045
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 |
|---|---|---|---|---|
| 1.2768 | 1.0 | 2506 | 1.8614 | 66.7187 |
| 0.9077 | 2.0 | 5012 | 1.9243 | 66.6383 |
| 0.6044 | 3.0 | 7518 | 2.0816 | 71.3474 |
| 0.4112 | 4.0 | 10024 | 2.2672 | 69.5064 |
| 0.2788 | 5.0 | 12530 | 2.4746 | 65.6728 |
| 0.1736 | 6.0 | 15036 | 2.6423 | 68.1196 |
| 0.096 | 7.0 | 17542 | 2.7603 | 66.7897 |
| 0.0632 | 8.0 | 20048 | 2.9008 | 68.4746 |
| 0.046 | 9.0 | 22554 | 3.0145 | 69.8850 |
| 0.0338 | 10.0 | 25060 | 3.0977 | 66.8749 |
| 0.0203 | 11.0 | 27566 | 3.1614 | 67.5186 |
| 0.0207 | 12.0 | 30072 | 3.2117 | 65.2847 |
| 0.011 | 13.0 | 32578 | 3.3028 | 66.4253 |
| 0.007 | 14.0 | 35084 | 3.3854 | 68.1102 |
| 0.0071 | 15.0 | 37590 | 3.3962 | 66.8702 |
| 0.0041 | 16.0 | 40096 | 3.4312 | 66.8323 |
| 0.0043 | 17.0 | 42602 | 3.4244 | 66.5294 |
| 0.0036 | 18.0 | 45108 | 3.4340 | 66.8512 |
| 0.0019 | 19.0 | 47614 | 3.4810 | 67.6558 |
| 0.0003 | 20.0 | 50120 | 3.4853 | 66.7045 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1