whisper-small-mal / README.md
manjugeorge's picture
End of training
c7adc5a verified
metadata
language:
  - ml
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Small manju Mal
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: ml
          split: None
          args: 'config: ml, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 99.08045977011494

Whisper Small manju Mal

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7138
  • Wer: 99.0805

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: 6000

Training results

Training Loss Epoch Step Validation Loss Wer
0.0069 37.0370 1000 0.6246 145.0575
0.0025 74.0741 2000 0.6177 99.3103
0.0 111.1111 3000 0.6737 101.3793
0.0 148.1481 4000 0.6957 99.0805
0.0 185.1852 5000 0.7087 99.3103
0.0 222.2222 6000 0.7138 99.0805

Framework versions

  • Transformers 4.41.0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1