metadata

library_name: transformers
language:
  - en
license: apache-2.0
base_model: openai/whisper-tiny
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - chinese_english_AE_fa_overfitting
metrics:
  - wer
model-index:
  - name: Whisper tiny Chinese
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Chinese English 'AE' Phonemes
          type: chinese_english_AE_fa_overfitting
          args: 'config: default, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 15.33186382561842

Whisper tiny Chinese

This model is a fine-tuned version of openai/whisper-tiny on the Chinese English 'AE' Phonemes dataset. It achieves the following results on the evaluation set:

Loss: 0.3921
Wer: 15.3319

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: 2
eval_batch_size: 1
seed: 42
optimizer: Use OptimizerNames.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: 3000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0616	2.0080	1000	0.3729	14.4012
0.0007	4.0161	2000	0.3843	15.0869
0.0005	6.0241	3000	0.3921	15.3319

Framework versions

Transformers 4.48.3
Pytorch 2.6.0+cu124
Datasets 3.4.1
Tokenizers 0.21.1