search

Zain-Jiang / Speech-Editing-Toolkit

It's a repository for implementations of neural speech editing algorithms.
174 stars 19 forks source link

Is there a Chinese example #11

Open kelvinqin opened 9 months ago

kelvinqin commented 9 months ago

Dear Jiang,

https://github.com/Zain-Jiang/Speech-Editing-Toolkit/issues/9#issuecomment-1714841548

You mentioned the current package support Chinese also, not sure if you can share an example please? Thanks for your excellent work. Today, after spent whole day, and I finally finished the installation of MFA, and also successfully built a model by running "run_mfa_train_align.sh". I plan to build a Chinese model soon, any suitable dataset you will suggest? Like Aishell2?

Kelvin

tangzhimiao commented 8 months ago

Dear Jiang,

#9 (comment)

You mentioned the current package support Chinese also, not sure if you can share an example please? Thanks for your excellent work. Today, after spent whole day, and I finally finished the installation of MFA, and also successfully built a model by running "run_mfa_train_align.sh". I plan to build a Chinese model soon, any suitable dataset you will suggest? Like Aishell2?

Kelvin

hey ，bro

i'm try use myself-data train chinese model , how to build chinese phone and words？ I'm having some problems

Zain-Jiang commented 8 months ago

@kelvinqin @tangzhimiao Hello kelvinqin and tangzhimiao. To build a Chinese model, I suggested using Aishell2+Aishell3+aidatatang+ST-CMDS and more clean ASR datasets. I'm now trying to add an Chinese example and pre-trained ckpts, which will takes about one or two weeks. If you don't have enough time to wait the example, you can refer to this script for building chinese phone and words https://github.com/Zain-Jiang/Dict-TTS/blob/main/data_gen/tts/txt_processors/zh.py.