search

NaomiProject / Naomi

The Naomi Project is an open source, technology agnostic platform for developing always-on, voice-controlled applications!
https://projectnaomi.com/
MIT License
242 stars 49 forks source link

Support for Rhasspy Piper TTS #385

Closed aaronchantrill closed 1 month ago

aaronchantrill commented 1 year ago

Detailed Description

See https://github.com/rhasspy/piper

Context

It adds additional voices and supports a large number of languages (about 20).

Possible Implementation

Currently it has to be downloaded and compiled. Then a voice needs to be selected. It should be possible to select voices based on the language selected in Naomi, as the filename includes the locale code.

aaronchantrill commented 4 months ago

There is a library called Dimits that makes using piper much easier. Just pip install dimits then from dimits import Dimits and dt = Dimits("en_US-amy-low") and audio_file = td.text_2_audio_file(text, "speech_output", "."). The last command will generate an audio file and return the name that can just be passed to your audio playback method.

aaronchantrill commented 1 month ago

I have created a "Piper-TTS" plugin and added it to NPE.