whisper-small-it / README.md
nefasto's picture
update model card README.md
03191a0
metadata
language:
  - it
license: apache-2.0
tags:
  - whisper-event
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Small Italian
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: mozilla-foundation/common_voice_11_0 it
          type: mozilla-foundation/common_voice_11_0
          config: it
          split: test
          args: it
        metrics:
          - name: Wer
            type: wer
            value: 12.303981501169467

Whisper Small Italian

This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_11_0 it dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2534
  • Wer: 12.3040

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: 8e-06
  • train_batch_size: 64
  • eval_batch_size: 32
  • 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: 6000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.2737 2.01 1000 0.2728 13.4097
0.1536 4.02 2000 0.2611 12.9897
0.0905 6.03 3000 0.2686 12.9273
0.1301 8.04 4000 0.2534 12.3040
0.096 10.05 5000 0.2727 12.6130
0.0604 12.06 6000 0.2698 12.5027

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.0+cu117
  • Datasets 2.7.1.dev0
  • Tokenizers 0.13.2