WakeWord AutoSynth Dataset

submit a .json file to ww_configs -> get a dataset under synth_data

to add a new voice submit a OPM plugin config to tts_voices

this automated pipeline is built with github workflows and OPM

Plugins

Only good quality plugins have been included by default

English currently outputs 180 unique samples for each wake word

• ovos-tts-plugin-mimic3
• ovos-tts-plugin-deepponies
• ovos-tts-plugin-privox
• ovos-tts-plugin-google-tx