metadata
library_name: transformers
language:
- ar
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
- generated_from_trainer
datasets:
- zw
metrics:
- wer
model-index:
- name: whisper-large-zwksa
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: zwksa
type: zw
config: ar
split: test
args: 'config: hi, split: test'
metrics:
- name: Wer
type: wer
value: 33.97781939701076
whisper-large-zwksa
This model is a fine-tuned version of openai/whisper-large-v3 on the zwksa dataset. It achieves the following results on the evaluation set:
- Loss: 0.2123
- Wer: 33.9778
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: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- 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 |
|---|---|---|---|---|
| 0.1846 | 0.4119 | 1000 | 0.2726 | 37.3784 |
| 0.159 | 0.8239 | 2000 | 0.2398 | 35.0722 |
| 0.1049 | 1.2358 | 3000 | 0.2214 | 33.1571 |
| 0.0751 | 1.6478 | 4000 | 0.2123 | 33.9778 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.3.1+cu121
- Datasets 3.2.0
- Tokenizers 0.19.1