README.md

metadata

license: apache-2.0
metrics:
  - wer
model-index:
  - name: openai/whisper-small-en
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: myst-test
          type: asr
          config: en
          split: test
        metrics:
          - type: wer
            value: 9.11
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: cslu_scripted
          type: asr
          config: en
          split: test
        metrics:
          - type: wer
            value: 33.85
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: cslu_spontaneous
          type: asr
          config: en
          split: test
        metrics:
          - type: wer
            value: 28.47
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: librispeech
          type: asr
          config: en
          split: testclean
        metrics:
          - type: wer
            value: 4.18
            name: WER

openai/whisper-small-en

This model is a fine-tuned version of openai/whisper-small-en on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.26971688866615295
• Wer: 8.508066331024994

Training and evaluation data

• Training data: Myst Train (125 hours)
• Evaluation data: Myst Dev (20.9 hours)

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 64
• eval_batch_size: 32
• 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: 10000
• converged_after: 1000