Mantella is a Skyrim and Fallout 4 mod which allows you to naturally speak to NPCs using Whisper (speech-to-text), LLMs (text generation), and xVASynth / XTTS (text-to-speech).
It would be nice to be able to use the memories or the discussion to generate an AI picture prompt.
I'm not sure if this should be considered outside or inside of Mantella.
Currently, the memories are in text, but if they move to chromadb, they'll need to be extracted.
In a feature complete function, the pic would be generated and used as background during the next fast travel / load time.
It would be nice to be able to use the memories or the discussion to generate an AI picture prompt.
I'm not sure if this should be considered outside or inside of Mantella. Currently, the memories are in text, but if they move to chromadb, they'll need to be extracted.
In a feature complete function, the pic would be generated and used as background during the next fast travel / load time.