Voice Activity Detection (VAD) AudioWorklet
-
Updated
Jun 10, 2024 - JavaScript
Voice Activity Detection (VAD) AudioWorklet
A real-time voice AI web app using Google Gemini Live API. Features speech-to-text, text-to-speech, and interruptible conversations, restricted to Revolt Motors topics. Built with Node.js, Express, WebSocket, and Web Audio API for smooth low-latency voice interactions.
🏦 VBank: Voice-Activated Banking Platform 🎙️ Talk to Your Bank Perform banking tasks like checking balance, transferring money, and viewing transactions using natural voice commands. 🧠 Smart & Secure 🔐 Biometric authentication (voice + face) 🧬 Liveness detection with OTP 🛡️ JWT-based session security
Local-first VAD, barge-in, and turn-taking primitives for interruptible voice agents.
Voice-enabled interactive AI avatar with a PixiJS frontend and Python/FastAPI/RAG backend that provides real-time voice activity detection and processing capabilities for rich multimedia experiences.
Full-stack recruitment platform with multi-agent LLM resume analysis, interview orchestration, and voice/face proctoring.
Pure-JS voice activity detection (SileroVAD v5) — no WASM, no ONNX Runtime, works on iOS Safari
A browser-based prototype that improves pause detection in voice AI interviews using adaptive VAD and turn-taking logic.
Add a description, image, and links to the vad topic page so that developers can more easily learn about it.
To associate your repository with the vad topic, visit your repo's landing page and select "manage topics."