metadata
language:
- fa
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_17
metrics:
- wer
model-index:
- name: Whisper Small FineTune Persian
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 17
type: mozilla-foundation/common_voice_17
config: fa
split: None
args: 'config: fa, split: test'
metrics:
- name: Wer
type: wer
value: 37.11215067988707
Whisper Small FineTune Persian
This model is a fine-tuned version of openai/whisper-small on the Common Voice 17 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3613
- Wer: 37.1122
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: 16
- 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: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2961 | 0.41 | 1000 | 0.4811 | 46.8883 |
| 0.2096 | 0.81 | 2000 | 0.4041 | 44.2752 |
| 0.1092 | 1.22 | 3000 | 0.3742 | 38.1105 |
| 0.0974 | 1.62 | 4000 | 0.3613 | 37.1122 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2