AIwaifu

Why we included this project

If you have watched neuro-sama and wished you could run an always-on streaming companion yourself, this project is a way in. It combines an open language model with text-to-speech, voice conversion, and live character animation into a single personality you can chat with. You supply your own model weights, define the persona and datasets, and the code handles the real-time pipeline around it, including translating replies and driving a VTube Studio avatar. Because every piece is open, you can swap the language model, change the voice, or reshape the character without hitting a proprietary API wall. It is a practical starting point for hobbyists who want a working end-to-end vocal chatbot to tinker with rather than assemble each component by hand.