Hey folks! In a long-awaited followup to my blog post on running Piper TTS, here’s a post describing how to bootstrap a FasterWhisper STT server. It doesn’t use containers deliberately, as many folks aren’t comfortable with them.
The script runs on any Debian-based OS but could be easily tweaked to run on other OSes. It sets up FasterWhisper using CPU inference and the medium.en model, but again, those settings are easily tweaked. For settings to apply to your server’s ~/.config/mycroft.mycroft.conf file, please see the plugin’s README: GitHub - OpenVoiceOS/ovos-stt-plugin-fasterwhisper: FasterWhisper for OpenVoiceOS
Please give it a spin and let me know if it works out for you! Happy to make any adjustments needed.
-Mike