metadata
license: apache-2.0
base_model: openai/whisper-medium
tags:
- generated_from_trainer
model-index:
- name: Whisper-medium-Arabic-phoneme-v2
results: []
Whisper-medium-Arabic-phoneme-v2
This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2289
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: 8
- 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
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.063 | 1.0 | 410 | 0.2223 |
| 0.0764 | 2.0 | 820 | 0.2359 |
| 0.0293 | 3.0 | 1230 | 0.1965 |
| 0.0174 | 4.0 | 1640 | 0.2038 |
| 0.0307 | 5.0 | 2050 | 0.2155 |
| 0.0075 | 6.0 | 2460 | 0.1972 |
| 0.0105 | 7.0 | 2870 | 0.2084 |
| 0.0035 | 8.0 | 3280 | 0.2173 |
| 0.0006 | 9.0 | 3690 | 0.2282 |
| 0.0006 | 10.0 | 4100 | 0.2289 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0