Not sure where to put this but I hope here will do.
I've been wanting to create a project for a while now - a completely offline speaking and listening AI assistant. However, I've run into a few problems....
I've been wanting to create a project for a while now - a completely offline speaking and listening AI assistant. However, I've run into a few problems....
- I wanted to use Whisper to convert user speech into text that the AI can then respond to. Firstly, I believe Whisper requires internet to run and secondly I have no clue on how I get the AI to read that data from Whisper
- My AI is stuck in the terminal. I don't know how to make any cool interface or anything, it runs only in the terminal.
- My AI is also slow. I believe I'm using a Qwen model with 8B parameters with 2Quanbits and its responding fairly slow, I wish to speed it up (please recommend me any AI that'd be faster.)
Statistics: Posted by haelstorm — Sat Aug 09, 2025 2:37 am