metadata
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: chinese-english-whisper-finetune-take2
results: []
chinese-english-whisper-finetune-take2
This model is a fine-tuned version of openai/whisper-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1626
- Wer: 82.7131
- Mer: 68.6675
- Cer: 28.2875
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: 0.0001
- 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: 50
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Mer | Cer |
|---|---|---|---|---|---|---|
| 0.1649 | 1.0811 | 200 | 1.2727 | 107.8031 | 70.1080 | 46.0846 |
| 0.1139 | 2.1622 | 400 | 1.2402 | 100.4802 | 71.9088 | 35.4459 |
| 0.0526 | 3.2432 | 600 | 1.2208 | 88.4754 | 69.0276 | 33.3300 |
| 0.0261 | 4.3243 | 800 | 1.1673 | 81.7527 | 66.9868 | 28.8709 |
| 0.0086 | 5.4054 | 1000 | 1.1626 | 82.7131 | 68.6675 | 28.2875 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1