Thinking about adding an option to the code of @Quackomatic for TTS just like glados_announcements has.
-
Thinking about adding an option to the code of @Quackomatic for TTS just like glados_announcements has. @glados_announcements@synthehai.z0ne.social
The idea is an optional setting in bot.cfg for a onnx model. If that's set then a MP3 is uploaded, if not nothing changes.
(TBF the code is in dire need of a refactor, but that would be a fun option, especially all the piper-tts stuff is already done basically in the glados announcements Repo) -
Thinking about adding an option to the code of @Quackomatic for TTS just like glados_announcements has. @glados_announcements@synthehai.z0ne.social
The idea is an optional setting in bot.cfg for a onnx model. If that's set then a MP3 is uploaded, if not nothing changes.
(TBF the code is in dire need of a refactor, but that would be a fun option, especially all the piper-tts stuff is already done basically in the glados announcements Repo)@fabos@der.cyberterror.ist @prettygood@socially.drinkingatmy.computer @EpicKitty@mk.catgirlsfor.science @puniko@mk.absturztau.be I would like to have your opinion on this idea.
If you don't mind
-
@fabos@der.cyberterror.ist @prettygood@socially.drinkingatmy.computer @EpicKitty@mk.catgirlsfor.science @puniko@mk.absturztau.be I would like to have your opinion on this idea.
If you don't mind -
@Erpel @puniko @EpicKitty @fabos forgive me for being drunk at midnight, but are you suggesting adding TTS for the existing Markov bots? If so I wonder if there's enough Kikuri voice lines to get some semblance of a voice print for this. A weeb can dream.
@puniko@mk.absturztau.be @EpicKitty@mk.catgirlsfor.science @fabos@der.cyberterror.ist @prettygood@socially.drinkingatmy.computer yes as an optional feature.
Dunno if there is a English KIKURI model, but at the moment I don't want to look into creating my own onnx model