How to use

(Suggestio) Pytho == 3.7

Cloe this repository

git cloe https://github.com/CjagCjegh/vits.git

Choose cleaers

• Fill "text_cleaers" i cofig.jso
• Edit text/symbols.py
• Remove uecessary imports from text/cleaers.py

Istall requiremets

pip istall -r requiremets.txt

Create datasets

Sigle speaker

"_speakers" should be 0 i cofig.jso

path/to/XXX.wav|trascript

• Example

dataset/001.wav|こんにちは。

Mutiple speakers

Speaker id should start from 0

path/to/XXX.wav|speaker id|trascript

• Example

dataset/001.wav|0|こんにちは。

Preprocess

If you have doe this, set "cleaed_text" to true i cofig.jso

# Sigle speaker
pytho preprocess.py --text_idex 1 --filelists path/to/filelist_trai.txt path/to/filelist_val.txt

# Mutiple speakers
pytho preprocess.py --text_idex 2 --filelists path/to/filelist_trai.txt path/to/filelist_val.txt

Build mootoic aligmet search

cd mootoic_alig
pytho setup.py build_ext --iplace
cd ..

Trai

# Sigle speaker
pytho trai.py -c <cofig> -m <folder>

# Mutiple speakers
pytho trai_ms.py -c <cofig> -m <folder>

Iferece

Olie

See iferece.ipyb

Offlie

See MoeGoe

Ruig i Docker

docker ru -itd --gpus all --ame "Cotaier ame" -e NVIDIA_DRIVER_CAPABILITIES=compute,utility -e NVIDIA_VISIBLE_DEVICES=all "Image ame"

How to use (Suggestion) Python == 3.7 Clone this repository git clone https://github.com/CjangCjengh