Skip to content

Latest commit

 

History

History

vits

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
__pycache__
__pycache__
 
 
configs
configs
 
 
filelists
filelists
 
 
monotonic_align
monotonic_align
 
 
resources
resources
 
 
text
text
 
 
README.md
README.md
 
 
attentions.py
attentions.py
 
 
commons.py
commons.py
 
 
data_utils.py
data_utils.py
 
 
inf.txt
inf.txt
 
 
inference.ipynb
inference.ipynb
 
 
libc-2.17.so
libc-2.17.so
 
 
losses.py
losses.py
 
 
mel_processing.py
mel_processing.py
 
 
mntts2_inference_multispk.py
mntts2_inference_multispk.py
 
 
models.py
models.py
 
 
modules.py
modules.py
 
 
preprocess.py
preprocess.py
 
 
requirements.txt
requirements.txt
 
 
test.txt
test.txt
 
 
train.py
train.py
 
 
train_ms.py
train_ms.py
 
 
train_mstet.py
train_mstet.py
 
 
transforms.py
transforms.py
 
 
utils.py
utils.py
 
 

README.md

VITS

Pre-requisites

  1. Python >= 3.6
  2. Clone this repository
  3. Install python requirements. Please refer requirements.txt
    1. You may need to install espeak first: apt-get install espeak
    2. For mult-speaker setting, download and extract the VCTK dataset, and downsample wav files to 22050 Hz. Then rename or create a link to the dataset folder: ln -s /path/to/MnTTS2/wavs_16bit MnTTS2
  4. Build Monotonic Alignment Search and run preprocessing.
# Cython-version Monotonoic Alignment Search
cd monotonic_align
python setup.py build_ext --inplace

# Preprocessing (g2p) for MnTTS2.
python preprocess.py --text_index 2 --filelists filelists/train.txt filelists/valid.txt filelists/test.txt

Training Exmaple

# MnTTS2
python train_ms.py -c configs/mntts2.json -m mntts2

Inference Example

python mntts2_inference_multispk.py