out
This model is a fine-tuned version of openai/whisper-small on the common_language dataset. It achieves the following results on the evaluation set:
- Loss: 0.6339
- Accuracy: 0.8898
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: 32
- eval_batch_size: 32
- seed: 0
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4103 | 1.0 | 694 | 1.1299 | 0.7471 |
| 0.5324 | 2.0 | 1388 | 0.6167 | 0.8495 |
| 0.3164 | 3.0 | 2082 | 0.5136 | 0.8679 |
| 0.1535 | 4.0 | 2776 | 0.5264 | 0.8758 |
| 0.1018 | 5.0 | 3470 | 0.5890 | 0.8767 |
| 0.0555 | 6.0 | 4164 | 0.6121 | 0.8772 |
| 0.031 | 7.0 | 4858 | 0.6624 | 0.8777 |
| 0.0264 | 8.0 | 5552 | 0.6172 | 0.8869 |
| 0.0079 | 9.0 | 6246 | 0.6225 | 0.8884 |
| 0.0058 | 10.0 | 6940 | 0.6339 | 0.8898 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.8.0+cu111
- Datasets 2.11.0
- Tokenizers 0.13.3
- Downloads last month
- 2
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.