Skip to content
VRCWizard edited this page Feb 26, 2023 · 6 revisions
  • STT method "Vosk"
    • Download models below or from the official website here (avoid 0.22 models, they have an issue with repeating "the" when nothing is spoken)
Recommended Model Download RAM Usage
vosk-model-en-us-0.42-gigaspeech High
vosk-model-en-us-daanzu-20200905 Medium
vosk-model-en-us-daanzu-20200905-lgraph Low
  • After downloading the model you wish to use, extract it and select the your model folder in Speech Provider > Local image
    • known bugs:
      • If you try using Vosk without having a model in the folder the program will crash, caused by System.AccessViolationException: 'Attempted to read or write protected memory. This is often an indication that other memory is corrupt.'
      • Crashing may also be noticed when stopping listening when using vosk

Need Help / Have Questions / Wanna make suggestions?

Donate

  • Leave me a Github Star ⭐ (it's free) or

Buy Me a Coffee at ko-fi.com

Clone this wiki locally