README.md

metadata

base_model: openai/whisper-small
datasets:
  - mozilla-foundation/common_voice_17_0
  - google/fleurs
  - facebook/multilingual_librispeech
language:
  - it
license: apache-2.0
metrics:
  - wer
tags:
  - whisper-event
  - generated_from_trainer
model-index:
  - name: Whisper Small Mixed-Italian
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: mozilla-foundation/common_voice_17_0 it
          type: mozilla-foundation/common_voice_17_0
          config: it
          split: test
          args: it
        metrics:
          - type: wer
            value: 10.587474512857398
            name: Wer
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: google/fleurs
          type: google/fleurs
          config: it_it
          split: test
        metrics:
          - type: wer
            value: 5.77
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: facebook/multilingual_librispeech
          type: facebook/multilingual_librispeech
          config: italian
          split: test
        metrics:
          - type: wer
            value: 13.52
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: facebook/voxpopuli
          type: facebook/voxpopuli
          config: it
          split: test
        metrics:
          - type: wer
            value: 25.87
            name: WER

Whisper Small Mixed-Italian

This model is a fine-tuned version of openai/whisper-small on the it datasets:

• mozilla-foundation/common_voice_17_0
• google/fleurs
• facebook/multilingual_librispeech

It achieves the following results on the evaluation set:

• Loss: 0.1909
• Wer: 10.5875

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: 64
• 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: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2213	0.2	1000	0.2407	13.4605
0.1582	0.4	2000	0.2143	12.2642
0.1913	0.6	3000	0.2022	11.2328
0.1538	0.8	4000	0.1951	11.1187
0.1286	1.0	5000	0.1909	10.5875

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1