metadata

library_name: transformers
language:
  - yue
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_17_0
metrics:
  - wer
model-index:
  - name: Whisper Small Canontese X v2
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 16.1 and 17.0
          type: mozilla-foundation/common_voice_17_0
          config: yue
          split: None
          args: 'config: zh-HK, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 54.825384904243336

Whisper Small Canontese X v2

This model is a fine-tuned version of openai/whisper-small on the Common Voice 16.1 and 17.0 dataset. It achieves the following results on the evaluation set:

Loss: 0.2264
Wer: 54.8254

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: 4
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: 3000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2578	0.6954	1000	0.2680	61.4345
0.0892	1.3908	2000	0.2376	57.3789
0.0295	2.0862	3000	0.2264	54.8254

Framework versions

Transformers 4.44.2
Pytorch 2.4.1+cu121
Datasets 3.0.0
Tokenizers 0.19.1