zh-CN-1-model / README.md
whucedar's picture
End of training
fd23073 verified
metadata
language:
  - zh
license: apache-2.0
base_model: openai/whisper-small
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - whucedar/retrain_jiaozhu
metrics:
  - wer
model-index:
  - name: zh-CN-1-model - whucedar
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: zh-CN
          type: whucedar/retrain_jiaozhu
          config: default
          split: None
          args: 'config: zh, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 90

zh-CN-1-model - whucedar

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

  • Loss: 0.0135
  • Wer: 90.0

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: 50
  • training_steps: 200
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0001 20.0 100 0.0135 90.0
0.0001 40.0 200 0.0135 90.0

Framework versions

  • Transformers 4.42.3
  • Pytorch 2.3.1+cu118
  • Datasets 2.20.0
  • Tokenizers 0.19.1