LEARN
Piper Plus voices
Some voices use the Piper Plus architecture: multilingual synthesis, prosody, and MB-iSTFT decoding. They carry the Piper+ badge in the catalogue and require the piper-plus runtime, not the standard piper binary.
Install piper-plus
The runtime ships on PyPI. The phonemizer NLTK data is fetched on first synthesis.
pip install piper-plusDownload a voice
The runtime resolves a model name and downloads its files. Catalogue Piper Plus voices stay available from their Hugging Face repositories too.
piper --download-model tsukuyomiGenerate a WAV file
Same interface as piper: text in, WAV file out. Multilingual voices cover their six languages in a single command.
piper --model tsukuyomi --text "Konnichiwa, kyou wa ii tenki desu ne." --output_file output.wavCatalogue models
Model pages carry the Piper+ badge and show the license. In-browser synthesis stays disabled for these voices: the site runtime only handles standard Piper.
A question about Piper Plus? The Discord server answers.