metadata
language:
- ar
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- zolfa
metrics:
- wer
model-index:
- name: Zolfa-raghadomar
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Zolfa Dataset
type: zolfa
args: 'config: ar, split: test'
metrics:
- name: Wer
type: wer
value: 14.285714285714285
Zolfa-raghadomar
This model is a fine-tuned version of openai/whisper-small on the Zolfa Dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.2369
- Wer: 14.2857
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: 4
- 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: 5
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0672 | 0.6993 | 100 | 0.2051 | 21.4286 |
| 0.0248 | 1.3986 | 200 | 0.2284 | 17.3469 |
| 0.0074 | 2.0979 | 300 | 0.2209 | 16.3265 |
| 0.0153 | 2.7972 | 400 | 0.2236 | 15.3061 |
| 0.0112 | 3.4965 | 500 | 0.2191 | 15.3061 |
| 0.0056 | 4.1958 | 600 | 0.2410 | 15.3061 |
| 0.0033 | 4.8951 | 700 | 0.2359 | 43.8776 |
| 0.0003 | 5.5944 | 800 | 0.2405 | 15.3061 |
| 0.0002 | 6.2937 | 900 | 0.2348 | 14.2857 |
| 0.0011 | 6.9930 | 1000 | 0.2369 | 14.2857 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1