whisper-tiny-dv / README.md
Maksimkrug's picture
End of training
3778a28
metadata
license: apache-2.0
base_model: openai/whisper-tiny
tags:
  - generated_from_trainer
datasets:
  - PolyAI/minds14
metrics:
  - wer
model-index:
  - name: whisper-tiny-dv
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: PolyAI/minds14
          type: PolyAI/minds14
          config: en-US
          split: train[450:]
          args: en-US
        metrics:
          - name: Wer
            type: wer
            value: 0.3400236127508855

whisper-tiny-dv

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: 0.6413
  • Wer Ortho: 0.3442
  • Wer: 0.3400

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: constant_with_warmup
  • lr_scheduler_warmup_steps: 50
  • training_steps: 500
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer Ortho Wer
0.3072 3.57 100 0.4919 0.3708 0.3595
0.0395 7.14 200 0.5423 0.3529 0.3430
0.0041 10.71 300 0.5857 0.3424 0.3359
0.0012 14.29 400 0.6185 0.3455 0.3394
0.0007 17.86 500 0.6413 0.3442 0.3400

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.1.0+cu121
  • Datasets 2.14.6
  • Tokenizers 0.14.1