|
--- |
|
language: |
|
- zh |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
base_model: openai/whisper-tiny |
|
datasets: |
|
- mozilla-foundation/common_voice_16_1 |
|
model-index: |
|
- name: Whisper Tiny chinese - VingeNie |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# Whisper Tiny chinese - VingeNie |
|
|
|
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Common Voice 16.1 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7976 |
|
- Cer Ortho: 35.9855 |
|
- Cer: 29.9636 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 32 |
|
- 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: 25 |
|
- training_steps: 1500 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Cer Ortho | Cer | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:| |
|
| 0.8364 | 1.0 | 300 | 0.8423 | 44.8697 | 32.3523 | |
|
| 0.5618 | 2.0 | 600 | 0.7816 | 44.3497 | 31.3998 | |
|
| 0.3559 | 3.0 | 900 | 0.7747 | 41.4869 | 30.1052 | |
|
| 0.2016 | 4.0 | 1200 | 0.7828 | 37.2903 | 30.2107 | |
|
| 0.0953 | 5.0 | 1500 | 0.7976 | 35.9855 | 29.9636 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.1 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.19.0 |
|
- Tokenizers 0.19.1 |
|
|
