|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- wer |
|
|
model-index: |
|
|
- name: Japanese_Fined_Tuned_Whisper_Model |
|
|
results: [] |
|
|
datasets: |
|
|
- mozilla-foundation/common_voice_11_0 |
|
|
language: |
|
|
- ja |
|
|
--- |
|
|
|
|
|
# Japanese_Fined_Tuned_Whisper_Model |
|
|
|
|
|
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Common Voice dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.780524 |
|
|
- Wer: 301.625840 |
|
|
|
|
|
## Model description |
|
|
|
|
|
The tiny Whisper model is fine-tuned on Japanese speech samples from the Common Voice dataset, based on which users can perform Automatic Speech Recognition in real time in Japanese. |
|
|
|
|
|
### Training hyperparameters |
|
|
|
|
|
The following hyperparameters were used during training: |
|
|
- learning_rate: 1e-05 |
|
|
- train_batch_size: 16 |
|
|
- 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: 4000 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Step | Validation Loss | Wer | |
|
|
|:-------------:|:-----:|:-----------------:|:----------:| |
|
|
| 0.4766 | 1000 | 0.739755 | 140.797746 | |
|
|
| 0.3136 | 2000 | 0.720202 | 190.830262 | |
|
|
| 0.1756 | 3000 | 0.773473 | 215.47997 | |
|
|
| 0.1472 | 4000 | 0.780524 | 301.625840 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.27.0.dev0 |
|
|
- Pytorch 1.13.1+cu116 |
|
|
- Datasets 2.10.1 |
|
|
- Tokenizers 0.13.2 |
