metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: openai/whisper-medium
results: []
openai/whisper-medium
This model is a fine-tuned version of openai/whisper-medium on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6150
- Wer: 20.4979
- Cer: 8.9847
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-06
- train_batch_size: 32
- eval_batch_size: 16
- 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.1041 | 0.2 | 1000 | 0.5133 | 22.6146 | 9.7131 |
| 0.074 | 1.11 | 2000 | 0.5532 | 21.6166 | 9.4196 |
| 0.0796 | 2.02 | 3000 | 0.6025 | 21.3314 | 9.2435 |
| 0.0422 | 2.22 | 4000 | 0.6029 | 20.7392 | 9.0274 |
| 0.0517 | 3.13 | 5000 | 0.6150 | 20.4979 | 8.9847 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2