Files
jarvis-models/tts/vits/README.md
2024-03-19 17:33:09 +08:00

69 lines
1.5 KiB
Markdown

# How to use
(Suggestion) Python == 3.7
## Clone this repository
```sh
git clone https://github.com/CjangCjengh/vits.git
```
## Choose cleaners
- Fill "text_cleaners" in config.json
- Edit text/symbols.py
- Remove unnecessary imports from text/cleaners.py
## Install requirements
```sh
pip install -r requirements.txt
```
## Create datasets
### Single speaker
"n_speakers" should be 0 in config.json
```
path/to/XXX.wav|transcript
```
- Example
```
dataset/001.wav|こんにちは。
```
### Mutiple speakers
Speaker id should start from 0
```
path/to/XXX.wav|speaker id|transcript
```
- Example
```
dataset/001.wav|0|こんにちは。
```
## Preprocess
If you have done this, set "cleaned_text" to true in config.json
```sh
# Single speaker
python preprocess.py --text_index 1 --filelists path/to/filelist_train.txt path/to/filelist_val.txt
# Mutiple speakers
python preprocess.py --text_index 2 --filelists path/to/filelist_train.txt path/to/filelist_val.txt
```
## Build monotonic alignment search
```sh
cd monotonic_align
python setup.py build_ext --inplace
cd ..
```
## Train
```sh
# Single speaker
python train.py -c <config> -m <folder>
# Mutiple speakers
python train_ms.py -c <config> -m <folder>
```
## Inference
### Online
See [inference.ipynb](inference.ipynb)
### Offline
See [MoeGoe](https://github.com/CjangCjengh/MoeGoe)
# Running in Docker
```sh
docker run -itd --gpus all --name "Container name" -e NVIDIA_DRIVER_CAPABILITIES=compute,utility -e NVIDIA_VISIBLE_DEVICES=all "Image name"
```