Whisper Tiny Optuna Akan
This model is a fine-tuned version of openai/whisper-tiny on the Speech Data Ghana UG - Ghanaian Multilingual Sample Data dataset. It achieves the following results on the evaluation set:
- Loss: 0.8282
- Wer: 65.8754
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.6121 | 2.9412 | 100 | 1.4449 | 94.9802 |
| 0.7595 | 5.8824 | 200 | 0.9033 | 71.6370 |
| 0.5632 | 8.8235 | 300 | 0.8282 | 65.8754 |
| 0.4631 | 11.7647 | 400 | 0.8235 | 67.1365 |
| 0.432 | 14.7059 | 500 | 0.8213 | 68.5460 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 15
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.
Model tree for nyarkssss/whisper-tiny-optuna
Base model
openai/whisper-tinyEvaluation results
- Wer on Speech Data Ghana UG - Ghanaian Multilingual Sample Dataself-reported65.875