README.md

metadata

language:
  - ar
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - Arbi-Houssem/Tunisian_dataset_STT-TTS30s_filtred1.0
metrics:
  - wer
model-index:
  - name: Whisper Tunisien
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Tunisian_dataset_STT-TTS30s_filtred1.0
          type: Arbi-Houssem/Tunisian_dataset_STT-TTS30s_filtred1.0
          args: 'config: ar, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 103.17460317460319

Whisper Tunisien

This model is a fine-tuned version of openai/whisper-small on the Tunisian_dataset_STT-TTS30s_filtred1.0 dataset. It achieves the following results on the evaluation set:

• Loss: 3.7250
• Wer: 103.1746

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: 8
• 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: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1045	8.4746	500	3.3315	102.7898
0.0043	16.9492	1000	3.7250	103.1746

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1