alikia2x/agnlash
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 1 Wiki Activity
Speech Recognition for Uyghur using deep learning
41 Commits 1 Branch 1 Tag 6.7 MiB
Python 100%
master
Go to file
alikia2x 496efa87bc
feature: implement training
2024-10-20 05:01:15 +08:00
assets ref: re-organize project structure 2024-10-20 03:36:48 +08:00
test ref: re-organize project structure 2024-10-20 03:36:48 +08:00
.gitignore feature: implement training 2024-10-20 05:01:15 +08:00
commonvoice.py feature: implement training 2024-10-20 05:01:15 +08:00
data.py feature: implement training 2024-10-20 05:01:15 +08:00
README.md ref: re-organize project structure 2024-10-20 03:36:48 +08:00
requirements.txt ref: re-organize project structure 2024-10-20 03:36:48 +08:00
tonu.py ref: re-organize project structure 2024-10-20 03:36:48 +08:00
train.py feature: implement training 2024-10-20 05:01:15 +08:00
UModel.py ref: re-organize project structure 2024-10-20 03:36:48 +08:00
uyghur.py Add files via upload 2021-06-14 16:56:51 +09:00

README.md

Agnlash

A ASR(Automatic Speech Recognition) model for Uyghur language.

This project is forked from uyghur-asr-ctc.

The Anglash is fine-tuned on the CommonVoice dataset which contains 313 hours of data.

The original project uses A free Uyghur speech database Released by CSLT@Tsinghua University & Xinjiang University. This dataset contains 22.45 hours of data.