metadata
library_name: transformers
language:
- ar
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- DrAliGomaa/ar-eg-dataset
metrics:
- wer
model-index:
- name: whisper-small-ar-test-draligomaa-dataset
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: draligomaa-dataset
type: DrAliGomaa/ar-eg-dataset
config: default
split: train[500:]
args: default
metrics:
- name: Wer
type: wer
value: 32.773109243697476
whisper-small-ar-test-draligomaa-dataset
This model is a fine-tuned version of openai/whisper-small on the draligomaa-dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.6208
- Wer Ortho: 32.7731
- Wer: 32.7731
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: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 20
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.006 | 6.25 | 100 | 0.5061 | 32.8571 | 32.8571 |
| 0.0027 | 12.5 | 200 | 0.5620 | 32.8571 | 32.8571 |
| 0.0009 | 18.75 | 300 | 0.5810 | 33.3193 | 33.3193 |
| 0.0006 | 25.0 | 400 | 0.5959 | 32.1429 | 32.1429 |
| 0.0004 | 31.25 | 500 | 0.6208 | 32.7731 | 32.7731 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1