check-malayalam
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1187
- Wer: 54.0682
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: 100
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.7411 | 0.46 | 100 | 0.5833 | 100.1244 |
| 0.2408 | 0.93 | 200 | 0.2089 | 72.5056 |
| 0.1299 | 1.39 | 300 | 0.1441 | 61.8811 |
| 0.1046 | 1.85 | 400 | 0.1250 | 55.8845 |
| 0.0763 | 2.31 | 500 | 0.1187 | 54.0682 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.1+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1