metadata
library_name: transformers
language:
- ar
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper base AR - BA
results: []
Whisper base AR - BA
This model is a fine-tuned version of openai/whisper-base on the quran-ayat-speech-to-text dataset. It achieves the following results on the evaluation set:
- Loss: 0.1094
- Wer: 0.3085
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.6308 | 0.9987 | 595 | 0.1030 | 0.2829 |
| 2.653 | 1.9987 | 1190 | 0.1033 | 0.2764 |
| 2.0618 | 2.9987 | 1785 | 0.1053 | 0.2763 |
| 1.6073 | 3.9987 | 2380 | 0.1087 | 0.3029 |
| 1.5376 | 4.9987 | 2975 | 0.1034 | 0.3090 |
| 1.2236 | 5.9987 | 3570 | 0.1001 | 0.2902 |
| 1.0811 | 6.9987 | 4165 | 0.1010 | 0.2768 |
| 1.0269 | 7.9987 | 4760 | 0.1003 | 0.3130 |
| 0.9971 | 8.9987 | 5355 | 0.0991 | 0.2864 |
| 0.9032 | 9.9987 | 5950 | 0.0996 | 0.3194 |
| 0.7539 | 10.9987 | 6545 | 0.1008 | 0.2734 |
| 0.7127 | 11.9987 | 7140 | 0.0985 | 0.2832 |
| 0.7146 | 12.9987 | 7735 | 0.0993 | 0.3050 |
| 0.6478 | 13.9987 | 8330 | 0.0994 | 0.2921 |
| 0.624 | 14.9987 | 8925 | 0.0997 | 0.2841 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1