Realtime voice-enabled AI assistant that can engage in natural conversations

A PyPI package that wraps Whisper.cpp for speech-to-text (STT) transcription, compatible with FastRTC

Real-time conversational AI with voice cloning and emotion detection. Analyzes conversation context to deliver dramatically expressive responses using your cloned voice. Built with FastRTC and Chatterbox TTS for natural, emotionally-aware voice interactions.

A turn-taking detection system for conversational AI built on top of FastRTC.

FastRTC implementation across few use cases. This is a basic initial version created only to demonstrate capabilities.

Real Time ASR Interface (HISI)

A FastAPI-based backend service for my portfolio website featuring voice-enabled Retrieval Augmented Generation (RAG) capabilities.

Build a real-time communication using Fasttrc, ollama, kokoro and etc