README.md

metadata

base_model: openai/whisper-tiny
datasets:
  - fleurs
language:
  - en
license: apache-2.0
metrics:
  - wer
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
model-index:
  - name: Whisper Tiny English - Chee Li
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: Google Fleurs
          type: fleurs
          config: en_us
          split: None
          args: 'config: en split: test'
        metrics:
          - type: wer
            value: 18.418851087562743
            name: Wer

Whisper Tiny English - Chee Li

This model is a fine-tuned version of openai/whisper-tiny on the Google Fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 0.6666
• Wer: 18.4189

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: 16
• 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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0468	5.3191	1000	0.5352	17.3592
0.0042	10.6383	2000	0.6165	18.6350
0.002	15.9574	3000	0.6532	18.3143
0.0016	21.2766	4000	0.6666	18.4189

Framework versions

• Transformers 4.43.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1