Voice AI
AIVoice
A local-first AI voice assistant: natural voice conversations via speech-to-text, LLM processing, and text-to-speech. FastAPI backend, WebSocket streaming, multiple voice personalities.
AIVoice is a full-stack voice assistant that captures audio from your microphone, transcribes with Whisper (faster-whisper), generates responses via an LLM (LM Studio), and speaks back using Piper TTS. It features push-to-talk, 6+ voice personalities, a modern web UI, and runs entirely locally for privacy and low latency.
Stack: FastAPI, faster-whisper, Piper TTS, WebSockets, LM Studio (OpenAI-compatible API), vanilla JS and Tailwind on the frontend.