|
--- |
|
language: |
|
- uz |
|
license: apache-2.0 |
|
tags: |
|
- automatic-speech-recognition |
|
- mozilla-foundation/common_voice_10_0 |
|
- generated_from_trainer |
|
- AIRI_UZ |
|
datasets: |
|
- common_voice_10_0 |
|
model-index: |
|
- name: uzbek_stt |
|
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. --> |
|
|
|
## Oyqiz jamoasi a'zolari tomonidan qilingan STT ning eng kichik versiyasi. Bu yaxshi versiya emas faqat 15ta epoch o'qitilgan! |
|
### Foziljon To'lqinov, Shaxboz Zohidov, Abduraxim Jabborov, Yahyoxon Rahimov |
|
|
|
|
|
|
|
Bu model [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) va MOZILLA-FOUNDATION/COMMON_VOICE_10_0 - UZ dataseti bilan 15ta epoxta o'qitilgan. |
|
O'qitish natijalari: |
|
- Xatolik: 0.5763 |
|
- So'z xatoligi: 0.4502 |
|
|
|
|
|
### O'qitish giperparameterlari |
|
|
|
O'qitish uchun ishlatilgan giperparameterlar: |
|
- learning_rate: 0.0003 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- total_train_batch_size: 8 |
|
- total_eval_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 15.0 |
|
- mixed_precision_training: Native AMP |
|
|
|
### O'qitish natijalari |
|
|
|
| Xatolik | Epox | Qadam | Tasdiq xatoligi | SXD | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------:| |
|
| 0.4736 | 3.4 | 10000 | 0.6247 | 0.6362 | |
|
| 0.3392 | 6.8 | 20000 | 0.7254 | 0.5605 | |
|
| 0.2085 | 10.19 | 30000 | 0.5465 | 0.5097 | |
|
| 0.1387 | 13.59 | 40000 | 0.5984 | 0.4632 | |
