Support Ukrainian VITS models from coqui-ai/TTS

[csukuangfj]

CC @magicse

You can find the model at
https://github.com/k2-fsa/sherpa-onnx/releases/tag/tts-models

Usage of the model

Method 1

git clone https://github.com/k2-fsa/sherpa-onnx
cd sherpa-onnx
mkdir build
cd build
cmake ..
make -j

wget https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/vits-coqui-uk-mai.tar.bz2
tar xf vits-coqui-uk-mai.tar.bz2

./bin/sherpa-onnx-offline-tts-play
  --vits-model=./vits-coqui-uk-mai/model.onnx \
  --vits-tokens=./vits-coqui-uk-mai/tokens.txt \
  --num-threads=1 \
  --output-filename=./abc.wav \
  --debug=1 \
  "Сьогодні, як завжди, люди діляться на дві групи: раби і вільні люди. Той, хто не має двох третин свого дня для себе, є рабом, ким би він не був: державним діячем, бізнесменом, чиновником чи вченим. "

Method 2

You can visit the following huggingface space
https://huggingface.co/spaces/k2-fsa/text-to-speech

Method 3

You can download the following exe for Windows

https://github.com/k2-fsa/sherpa-onnx/releases/download/tts-models/sherpa-onnx-non-streaming-tts-x64-v1.8.11.exe

[csukuangfj]

A screenshot of the model directory is given below.

You can find how the model is exported to sherpa-onnx by looking at the two scripts in the model directory.