Skip to content
/ nvidia-nemo-jasper-quartznet-asr-vietnamese Public

Nhận dạng giọng nói Tiếng Việt sử dụng model Quartznet (Nvidia) + flask demo

1 star 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

dangvansam/nvidia-nemo-jasper-quartznet-asr-vietnamese

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

38 Commits
NeMo
NeMo
 
 
__pycache__
__pycache__
 
 
apex
apex
 
 
audio_test
audio_test
 
 
config
config
 
 
converted2phone_simple
converted2phone_simple
 
 
converted2phone_vphon
converted2phone_vphon
 
 
data
data
 
 
g2pNp2g_simple
g2pNp2g_simple
 
 
nemo
nemo
 
 
pretrained_quartznet15x5
pretrained_quartznet15x5
 
 
quartznet12x1_12042021_finetune_from_15012021
quartznet12x1_12042021_finetune_from_15012021
 
 
quartznet12x1_15012021
quartznet12x1_15012021
 
 
quartznet12x1_char_24022021
quartznet12x1_char_24022021
 
 
quartznet12x1_char_all_data
quartznet12x1_char_all_data
 
 
static
static
 
 
templates
templates
 
 
text_ori
text_ori
 
 
wandb
wandb
 
 
README.md
README.md
 
 
compare_text.py
compare_text.py
 
 
eval.py
eval.py
 
 
eval_gridsearch.py
eval_gridsearch.py
 
 
flask_upload_only_vn.py
flask_upload_only_vn.py
 
 
flask_upload_record_en.py
flask_upload_record_en.py
 
 
flask_upload_record_vn.py
flask_upload_record_vn.py
 
 
get_text.py
get_text.py
 
 
git_push.sh
git_push.sh
 
 
infer.py
infer.py
 
 
load_label.py
load_label.py
 
 
make_json_file.py
make_json_file.py
 
 
nemo-flask-web.PNG
nemo-flask-web.PNG
 
 
nohup.out
nohup.out
 
 
out_text.txt
out_text.txt
 
 
out_text_phoneme.txt
out_text_phoneme.txt
 
 
process_text_lm.py
process_text_lm.py
 
 
result.txt
result.txt
 
 
train.py
train.py
 
 
train_abc.py
train_abc.py
 
 
train_fintune.py
train_fintune.py
 
 
train_vivos.py
train_vivos.py
 
 

Repository files navigation

nvidia-nemo-jasper-quartznet-asr-vietnamese

Nhận dạng giọng nói Tiếng Việt sử dụng model Quartznet (Nvidia) + flask demo
Sử dụng đơn vị Âm vị (Phoneme) cho CTC loss và convert ngược lại Kí tự (Grapheme): https://github.com/dangvansam98/phoneme2grapheme-vietnamese
model trained with 50h dataset(vivos, fpt, ...)
alt-text

About

Nhận dạng giọng nói Tiếng Việt sử dụng model Quartznet (Nvidia) + flask demo

Topics

speech-recognition automatic-speech-recognition speech-to-text vietnamese-nlp vietnamese-language phoneme-conversion vietnamese-speech-recognition

Resources

Readme
Activity

Stars

1 star

Watchers

2 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages