README.md

metadata

library_name: transformers
language:
  - zh
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Small chinese Test
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: nan-tw
          split: test
          args: 'config: zh-tw, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 94.0629839958699

Whisper Small chinese Test

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.9213
• Wer: 94.0630

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• 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
0.1069	3.6364	1000	0.7541	99.3289
0.0117	7.2727	2000	0.8330	93.9597
0.0015	10.9091	3000	0.8627	94.7858
0.0004	14.5455	4000	0.9036	93.3918
0.0002	18.1818	5000	0.9213	94.0630

Framework versions

• Transformers 4.46.3
• Pytorch 2.4.0+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3