metadata
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-tiny-si
results: []
whisper-tiny-si
This model is a fine-tuned version of openai/whisper-tiny on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6177
- Wer: 437.6471
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: 5
- training_steps: 40
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 1.43 | 10 | 2.6314 | 370.5882 |
| No log | 2.86 | 20 | 1.9292 | 532.9412 |
| 2.6969 | 4.29 | 30 | 1.6605 | 376.4706 |
| 2.6969 | 5.71 | 40 | 1.6177 | 437.6471 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0