english / README.md
mn720's picture
End of training
c76d687 verified
metadata
language:
  - en
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_15_0
metrics:
  - wer
model-index:
  - name: Whisper Small Luganda
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 15.0
          type: mozilla-foundation/common_voice_15_0
          config: lg
          split: validation
          args: 'config: lu, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 40.4513375311668

Whisper Small Luganda

This model is a fine-tuned version of openai/whisper-small on the Common Voice 15.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3827
  • Wer: 40.4513

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
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 4000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.6718 0.1129 500 0.7164 66.9178
0.4723 0.2258 1000 0.5436 54.4267
0.4208 0.3388 1500 0.4769 49.0925
0.3881 0.4517 2000 0.4404 45.3112
0.3739 0.5646 2500 0.4167 43.9251
0.3387 0.6775 3000 0.3993 41.2913
0.3405 0.7904 3500 0.3886 41.2099
0.3089 0.9033 4000 0.3827 40.4513

Framework versions

  • Transformers 4.40.0
  • Pytorch 2.2.2+cu118
  • Datasets 2.19.0
  • Tokenizers 0.19.1