metadata
library_name: transformers
language:
- ar
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Whisper tiny AR - BH
results: []
Whisper tiny AR - BH
This model is a fine-tuned version of openai/whisper-tiny on the quran-ayat-speech-to-text dataset. It achieves the following results on the evaluation set:
- Loss: 0.0076
- Wer: 0.0861
- Cer: 0.0359
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: 5e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Use 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: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.008 | 1.0 | 235 | 0.0066 | 0.0804 | 0.0302 |
| 0.0062 | 2.0 | 470 | 0.0064 | 0.0793 | 0.0305 |
| 0.0066 | 3.0 | 705 | 0.0063 | 0.0753 | 0.0302 |
| 0.0038 | 4.0 | 940 | 0.0064 | 0.0767 | 0.0288 |
| 0.0037 | 5.0 | 1175 | 0.0067 | 0.0766 | 0.0296 |
| 0.0031 | 6.0 | 1410 | 0.0069 | 0.0751 | 0.0291 |
| 0.002 | 7.0 | 1645 | 0.0074 | 0.0809 | 0.0309 |
| 0.0013 | 8.0 | 1880 | 0.0077 | 0.0796 | 0.0297 |
| 0.0016 | 9.0 | 2115 | 0.0079 | 0.0796 | 0.0297 |
| 0.001 | 10.0 | 2350 | 0.0083 | 0.0795 | 0.0297 |
| 0.0005 | 11.0 | 2585 | 0.0085 | 0.0780 | 0.0290 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0