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-CN
split: test
args: zh-CN
metrics:
- name: Wer
type: wer
value: 72.67721353028581
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: 0.4019
- Wer: 72.6772
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: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.5179 | 2.02 | 1000 | 0.3333 | 72.9831 |
| 0.1273 | 4.04 | 2000 | 0.3562 | 73.9621 |
| 0.0163 | 6.06 | 3000 | 0.3790 | 73.9708 |
| 0.004 | 8.07 | 4000 | 0.3946 | 72.3626 |
| 0.025 | 11.0 | 5000 | 0.4019 | 72.6772 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2