README.md

metadata

license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - audiofolder
metrics:
  - wer
model-index:
  - name: whisper-small-ft_custom
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: audiofolder
          type: audiofolder
          config: default
          split: None
          args: default
        metrics:
          - name: Wer
            type: wer
            value: 80

whisper-small-ft_custom

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

• Loss: 1.9637
• Wer: 80.0

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-06
• 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: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.7875	35.71	500	2.2491	514.0
0.0077	71.43	1000	1.9637	80.0

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2