Description
You want a real-time voice assistant without sending your voice to the cloud. Local Voice AI is a private, low-latency voice assistant running on your own hardware, combining speech recognition, an LLM and speech synthesis in one app.
Built on LiveKit, llama.cpp, Kokoro and Nemotron STT, it picks a model stack for your hardware and even runs live on a Jetson Orin Nano.
Fully local:No cloud audio.
Low latency:Real-time talk.
Auto stack:Fits your hardware.
Edge:Jetson support.
Built on LiveKit, llama.cpp, Kokoro and Nemotron STT, it picks a model stack for your hardware and even runs live on a Jetson Orin Nano.
Features
Fully local:No cloud audio.
Low latency:Real-time talk.
Auto stack:Fits your hardware.
Edge:Jetson support.
