Whisper Large v2 Hindi
This model is a fine-tuned version of openai/whisper-large-v2 on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1915
- Wer: 12.4241
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0936 | 0.37 | 100 | 0.2300 | 16.2100 |
| 0.09 | 0.73 | 200 | 0.2117 | 14.4876 |
| 0.0415 | 1.1 | 300 | 0.2048 | 13.0832 |
| 0.0372 | 1.47 | 400 | 0.1951 | 12.5559 |
| 0.0307 | 1.84 | 500 | 0.1915 | 12.4241 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2