Conversation

Notices

1. Embed this notice
   Boiling Steam (boilingsteam@mastodon.cloud)'s status on Thursday, 04-Sep-2025 08:46:30 JST Boiling Steam

   Vibevoice: a Frontier Open-Source Text-To-Speech Model: https://microsoft.github.io/VibeVoice/
   #linux #tts #vibevoice #microsoft #model #ai

   • narcolepsy and alcoholism :flag: likes this.
   • Embed this notice
     narcolepsy and alcoholism :flag: (hj@shigusegubu.club)'s status on Thursday, 04-Sep-2025 08:47:05 JST narcolepsy and alcoholism :flag:

     in reply to
     @boilingsteam hope they add it to home assistant
   • Embed this notice
     narcolepsy and alcoholism :flag: (hj@shigusegubu.club)'s status on Thursday, 04-Sep-2025 18:28:54 JST narcolepsy and alcoholism :flag:

     in reply to
     @boilingsteam no CPU option? I currently (try to) run VOICEVOX for TTS.
   • Embed this notice
     Boiling Steam (boilingsteam@mastodon.cloud)'s status on Thursday, 04-Sep-2025 18:28:55 JST Boiling Steam

     in reply to

     • narcolepsy and alcoholism :flag:

     @hj need A GPU so I doubt it would be a good fit

   • Embed this notice
     narcolepsy and alcoholism :flag: (hj@shigusegubu.club)'s status on Thursday, 04-Sep-2025 21:51:40 JST narcolepsy and alcoholism :flag:

     in reply to
     @boilingsteam doesn't really matter given that HA caches TTS files, i.e. it computes same sentence once and then reuses same wav file
   • Embed this notice
     Boiling Steam (boilingsteam@mastodon.cloud)'s status on Thursday, 04-Sep-2025 21:51:41 JST Boiling Steam

     in reply to

     • narcolepsy and alcoholism :flag:

     @hj you can try CPU but it's super compute intensive so would be super slow on CPU