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 - BH
results: []
Whisper base AR - BH
This model is a fine-tuned version of openai/whisper-base on the quran-ayat-speech-to-text-segments dataset. It achieves the following results on the evaluation set:
- Loss: 0.0171
- Wer: 0.2407
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- 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 |
|---|---|---|---|---|
| 0.0105 | 2.6667 | 100 | 0.0154 | 0.2514 |
| 0.0021 | 5.3333 | 200 | 0.0171 | 0.2407 |
| 0.0022 | 8.0 | 300 | 0.0200 | 0.2475 |
| 0.0026 | 10.6667 | 400 | 0.0215 | 0.2847 |
| 0.0027 | 13.3333 | 500 | 0.0237 | 0.2929 |
| 0.0014 | 16.0 | 600 | 0.0216 | 0.2548 |
| 0.0001 | 18.6667 | 700 | 0.0210 | 0.2386 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.4.0
- Datasets 3.1.0
- Tokenizers 0.20.3