whisper-small-sq / README.md
kadriu's picture
End of training
c865fef
metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: whisper-small-sq
    results: []

whisper-small-sq

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

  • Loss: 0.2403
  • Wer: 0.2590

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: 8
  • 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: 8000

Training results

Training Loss Epoch Step Validation Loss Wer
0.3334 0.27 1000 0.3662 0.4138
0.2593 0.53 2000 0.3056 0.3329
0.2297 0.8 3000 0.2799 0.3052
0.1687 1.06 4000 0.2627 0.2798
0.1664 1.33 5000 0.2524 0.2724
0.1578 1.59 6000 0.2453 0.2652
0.1543 1.86 7000 0.2407 0.2626
0.1209 2.13 8000 0.2403 0.2590

Framework versions

  • Transformers 4.33.2
  • Pytorch 2.2.0.dev20230923
  • Datasets 2.14.5
  • Tokenizers 0.13.3