whisper-large-v2-finetuning-2
This model is a fine-tuned version of guilhermebastos96/whisper-large-v2-finetuning on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2251
- Wer: 11.8114
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: 16
- 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: 500
- training_steps: 6000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0724 | 0.5089 | 1000 | 0.2000 | 15.6703 |
| 0.0322 | 1.0178 | 2000 | 0.2156 | 12.0592 |
| 0.0398 | 1.5267 | 3000 | 0.2065 | 9.9843 |
| 0.0167 | 2.0356 | 4000 | 0.2091 | 10.5134 |
| 0.0107 | 2.5445 | 5000 | 0.2181 | 13.2453 |
| 0.0035 | 3.0534 | 6000 | 0.2251 | 11.8114 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.2.1
- Datasets 2.20.0
- Tokenizers 0.19.1