metadata
license: apache-2.0
base_model: openai/whisper-tiny.en
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisperFinetune
results: []
whisperFinetune
This model is a fine-tuned version of openai/whisper-tiny.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5788
- Wer: 23.0816
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: 0.001
- train_batch_size: 128
- 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: 100
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.837 | 0.2778 | 10 | 1.8599 | 36.0344 |
| 0.7523 | 0.5556 | 20 | 0.6735 | 27.1639 |
| 0.6525 | 0.8333 | 30 | 0.5461 | 23.2965 |
| 0.4156 | 1.1111 | 40 | 0.4989 | 22.3757 |
| 0.3478 | 1.3889 | 50 | 0.5054 | 22.4371 |
| 0.3294 | 1.6667 | 60 | 0.4953 | 22.4371 |
| 0.3547 | 1.9444 | 70 | 0.4787 | 20.1964 |
| 0.1265 | 2.2222 | 80 | 0.5288 | 22.2836 |
| 0.143 | 2.5 | 90 | 0.5582 | 29.0976 |
| 0.1498 | 2.7778 | 100 | 0.5788 | 23.0816 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.1.dev0
- Tokenizers 0.19.1