metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_11_0
metrics:
  - wer
model-index:
  - name: openai/whisper-small
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_11_0
          type: common_voice_11_0
          config: zh-TW
          split: test
          args: zh-TW
        metrics:
          - name: Wer
            type: wer
            value: 32.594792142530835

openai/whisper-small

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: 0.3250
Wer: 32.5948

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: 64
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 800
training_steps: 1000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3465	1.05	200	0.3499	41.9324
0.2137	2.1	400	0.2953	36.2951
0.1255	4.01	600	0.2927	33.7232
0.0509	5.06	800	0.3149	34.0566
0.0164	6.11	1000	0.3250	32.5948

Framework versions

Transformers 4.26.0.dev0
Pytorch 1.13.0+cu116
Datasets 2.7.1.dev0
Tokenizers 0.13.2