Scrya's picture
Update metadata with huggingface_hub
af3547b
metadata
language:
  - vi
license: apache-2.0
tags:
  - whisper-event
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Medium VI - CV - Augmented
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: vi
          split: test
          args: vi
        metrics:
          - type: wer
            value: 18.030269796007897
            name: Wer
          - type: wer
            value: 17.98
            name: WER
          - type: cer
            value: 8.31
            name: CER

Whisper Medium VI - CV - Augmented

This model is a fine-tuned version of openai/whisper-medium on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6613
  • Wer: 18.0303
  • Cer: 8.3095

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: 32
  • eval_batch_size: 16
  • 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: 5000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer Cer
0.0053 11.49 1000 0.5429 18.1290 8.4643
0.0021 22.99 2000 0.5916 18.8857 8.6538
0.0001 34.48 3000 0.6348 18.3374 8.4296
0.0001 45.98 4000 0.6508 17.9754 8.3149
0.0001 57.47 5000 0.6613 18.0303 8.3095

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.1+cu117
  • Datasets 2.7.1.dev0
  • Tokenizers 0.13.2