🎤💬 Full example of implementing ChatGPT's realtime voice from scratch with VAD + STT + LLM + TTS technology stack within almost one file!

Real-time voice agents with parallel async background sub-agents — conversations continue naturally while tasks run • Join the builders → https://discord.gg/mqxKaN3UKC

LiveKit voice app validation skill. Use when building, debugging, or declaring working any LiveKit voice agent, Agents UI app, or React/Next.js LiveKit project. Enforces evidence-based validation before reporting a session, token endpoint, worker, transcript, or end-to-end voice interaction as complete.

LiveKit Agents UI demo showing a voice AI assistant that schedules roof inspections using real-time voice interaction, visualizers, and booking workflow.

An AI-powered object detection system using YOLOv8 to identify and locate graffiti across various contexts including walls, buildings, over-bridges, vehicles, and other surfaces.

Real-time hand sign recognition using LSTM-based models for sequence detection from video frames.

A real-time (<500ms) voice AI concierge built with Next.js, FastAPI, and Gemini 2.5 Flash Lite. Features local RAG (ChromaDB) for policy retrieval, Tool Calling for live booking, and event-driven CRM logging to Google Sheets.

Traffyx-AI — Traffic Forecasting & Urban Mobility Intelligence System Applied machine learning system for traffic prediction, congestion analysis, and real-world spatiotemporal data modeling.

Example apps showcase what can be build with the Livepeer BYOC workflow.

Production-ready real-time voice AI pipeline integrating Twilio Media Streams, streaming ASR (Deepgram), LLM reasoning, and live analytics dashboard. Designed for ultra-low latency conversational intelligence in call center and healthcare environments.

VoxGuard is a real-time multimodal scam detection system for live calls, built with Gemini Live API, Rust WASM audio streaming, and psychological manipulation scoring.

High-performance async Python backend for real-time AI conversations with Quart, Supabase, and OpenAI.

Real-time face verification system using MediaPipe Face Mesh and landmark-based geometric feature extraction for improved accuracy and robustness.

Voice agent prototype for structured clinical interviewing, with VAD-based interruption handling, modular ASR/LLM/TTS backends, and dialogue workflow control.

Most AI tools help you after the call. PitchPulse helps you during it — guiding discovery and building your pitch in real time, so you can close while others guess.

A face recognition system implemented using Principal Component Analysis (PCA) and Artificial Neural Networks (ANN) that extracts eigenfaces for dimensionality reduction and performs identity classification using a neural network model.