whisper-tiny-en / README.md
Maimonator's picture
update model card README.md
8f29dd4
metadata
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - PolyAI/minds14
metrics:
  - wer
model-index:
  - name: whisper-tiny-en
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: PolyAI/minds14
          type: PolyAI/minds14
          config: en-US
          split: train
          args: en-US
        metrics:
          - name: Wer
            type: wer
            value: 0.3541912632821724

whisper-tiny-en

This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6875
  • Wer Ortho: 0.3745
  • Wer: 0.3542

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: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 500
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer Ortho Wer
0.6838 1.79 50 0.6522 0.4028 0.3613
0.2778 3.57 100 0.5727 0.3880 0.3589
0.1313 5.36 150 0.5870 0.3794 0.3501
0.0539 7.14 200 0.6080 0.3726 0.3471
0.022 8.93 250 0.6380 0.3745 0.3477
0.0095 10.71 300 0.6629 0.3843 0.3595
0.0049 12.5 350 0.6715 0.3819 0.3583
0.0036 14.29 400 0.6811 0.3825 0.3595
0.0032 16.07 450 0.6858 0.3757 0.3554
0.0029 17.86 500 0.6875 0.3745 0.3542

Framework versions

  • Transformers 4.29.0.dev0
  • Pytorch 2.0.0+cu117
  • Datasets 2.11.0
  • Tokenizers 0.13.3