|
--- |
|
language: |
|
- ko |
|
license: apache-2.0 |
|
base_model: openai/whisper-tiny |
|
tags: |
|
- hf-asr-leaderboard |
|
- generated_from_trainer |
|
datasets: |
|
- haseong8012/child-50k |
|
model-index: |
|
- name: whisper-small_child50K_timestretch_stepLR |
|
results: [] |
|
metrics: |
|
- wer |
|
- cer |
|
--- |
|
|
|
<!-- 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-small_child50K_timeStretch_stepLR |
|
|
|
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the child-50k dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0201 |
|
- Wer: 2.1061 |
|
- Cer: 0.8189 |
|
|
|
## 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: 1.25e-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: 5000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:| |
|
| 0.1004 | 0.18 | 500 | 0.0796 | 9.5440 | 4.0921 | |
|
| 0.0846 | 0.36 | 1000 | 0.0453 | 5.3319 | 2.3843 | |
|
| 0.0729 | 0.53 | 1500 | 0.0355 | 4.2849 | 1.7311 | |
|
| 0.0486 | 0.71 | 2000 | 0.0284 | 2.8701 | 1.2241 | |
|
| 0.045 | 0.89 | 2500 | 0.0261 | 3.6220 | 2.6581 | |
|
| 0.0206 | 1.07 | 3000 | 0.0207 | 2.0616 | 0.8263 | |
|
| 0.0264 | 1.24 | 3500 | 0.0219 | 2.0091 | 0.8275 | |
|
| 0.0304 | 1.42 | 4000 | 0.0205 | 1.7827 | 0.7207 | |
|
| 0.0224 | 1.6 | 4500 | 0.0233 | 2.3527 | 0.9527 | |
|
| 0.0215 | 1.78 | 5000 | 0.0201 | 2.1061 | 0.8189 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 1.12.1 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.14.1 |
