README.md

metadata

language:
  - ja
license: apache-2.0
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - Specific-Person-Voice
metrics:
  - wer
model-index:
  - name: Whisper Small ja - Tohio Uchiyama
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Special
          type: Specific-Person-Voice
          config: null
          split: None
          args: 'config: ja, split: train'
        metrics:
          - name: Wer
            type: wer
            value: 984.4311377245509

Whisper Small ja - Tohio Uchiyama

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

• Loss: 0.9332
• Wer: 984.4311

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5
• training_steps: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	2.0	10	1.1812	31.1377
No log	4.0	20	0.9332	984.4311

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.1+cpu
• Datasets 2.8.0
• Tokenizers 0.13.2