Original Repo - CLI Tool - Streaming API
python join.py audio1.wav audio2.mp3 audio3.mp4 audio4.ogg output.mp3
Docs:
StyleTTS 2 CLI Inference
Usage: cli.py [OPTIONS] TEXT OUTPUT
Options:
-a, --audio-sample PATH [required]
-p, --pth-path PATH [required]
-f, --file BOOLEAN Read from input text from file [default: False]
--help Show this message and exit.
Inference:
python cli.py "Hello, this is a test" -a 1.wav -p model.pth
Inference from a file:
python cli.py file.txt -a 1.wav -p model.pth -f
GPL due to Phonemizer (soon to be permissively licensed as soon as I find a permissively-licensed Phonemizer replacement)