metadata
language:
- fr
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: openai/whisper-small
results: []
openai/whisper-small
This model is a fine-tuned version of openai/whisper-small on the pphuc25/FrenchMed dataset. It achieves the following results on the evaluation set:
- Loss: 1.2271
- Wer: 40.1026
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 |
---|---|---|---|---|
0.8705 | 1.0 | 215 | 0.8589 | 164.5161 |
0.6103 | 2.0 | 430 | 0.8569 | 150.7331 |
0.3958 | 3.0 | 645 | 0.9063 | 72.4340 |
0.2181 | 4.0 | 860 | 0.9451 | 94.7947 |
0.137 | 5.0 | 1075 | 0.9824 | 113.8563 |
0.0827 | 6.0 | 1290 | 1.0459 | 68.6950 |
0.0429 | 7.0 | 1505 | 1.0825 | 86.7302 |
0.0259 | 8.0 | 1720 | 1.1261 | 113.0499 |
0.0215 | 9.0 | 1935 | 1.1081 | 44.6481 |
0.0161 | 10.0 | 2150 | 1.1658 | 64.5161 |
0.007 | 11.0 | 2365 | 1.1593 | 38.9296 |
0.0048 | 12.0 | 2580 | 1.1756 | 39.0029 |
0.0032 | 13.0 | 2795 | 1.1836 | 37.5367 |
0.002 | 14.0 | 3010 | 1.1991 | 38.9296 |
0.0038 | 15.0 | 3225 | 1.2011 | 37.7566 |
0.0011 | 16.0 | 3440 | 1.2117 | 40.7625 |
0.001 | 17.0 | 3655 | 1.2195 | 38.5630 |
0.0009 | 18.0 | 3870 | 1.2237 | 40.1760 |
0.0008 | 19.0 | 4085 | 1.2262 | 39.9560 |
0.0008 | 20.0 | 4300 | 1.2271 | 40.1026 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1