<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
whisper-base_trained
This model is a fine-tuned version of openai/whisper-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5384
- Wer: 150.0
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: 2
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- training_steps: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.5874 | 4.0 | 6 | 2.2372 | 150.0 |
| 1.1083 | 8.0 | 12 | 1.4557 | 150.0 |
| 0.6359 | 12.0 | 18 | 1.0874 | 150.0 |
| 0.2396 | 16.0 | 24 | 0.8668 | 200.0 |
| 0.056 | 20.0 | 30 | 0.7220 | 150.0 |
| 0.0147 | 24.0 | 36 | 0.6112 | 200.0 |
| 0.0055 | 28.0 | 42 | 0.5606 | 200.0 |
| 0.0037 | 32.0 | 48 | 0.5384 | 150.0 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
