|
|
--- |
|
|
license: cc-by-nc-sa-4.0 |
|
|
datasets: |
|
|
- Bingsu/KSS_Dataset |
|
|
language: |
|
|
- ko |
|
|
tags: |
|
|
- coqui-tts-pygoruut |
|
|
- coqui |
|
|
- tts |
|
|
- pygoruut |
|
|
--- |
|
|
# Coqui VITS KSS Korean TTS Model |
|
|
|
|
|
# Requirements |
|
|
|
|
|
- Python `3.10` |
|
|
- coqui-tts-pygoruut `0.27.4` |
|
|
|
|
|
# Install script |
|
|
|
|
|
```bash |
|
|
python3-alt.3-10 -m venv venv |
|
|
./venv/bin/pip3 install coqui-tts-pygoruut |
|
|
``` |
|
|
|
|
|
# Download the model |
|
|
|
|
|
```bash |
|
|
wget https://huggingface.co/neurlang/coqui-vits-kss-korean/resolve/main/best_model.pth -O /tmp/best_model.pth |
|
|
wget https://huggingface.co/neurlang/coqui-vits-kss-korean/resolve/main/config.json -O /tmp/config.json |
|
|
``` |
|
|
# Invoke TTS to File |
|
|
|
|
|
```bash |
|
|
./venv/bin/tts --text "μλ
νμΈμ μΈμ" --model_path /tmp/best_model.pth --config_path /tmp/config.json --out_path /tmp/synthesized.wav |
|
|
``` |
|
|
|
|
|
# Invoke TTS |
|
|
|
|
|
```bash |
|
|
./venv/bin/tts --text "μλ
νμΈμ μΈμ" --model_path /tmp/best_model.pth --config_path /tmp/config.json --out_path /tmp/synthesized.wav --pipe_out | aplay |
|
|
``` |