Python scripts to handle a two way voice conversation with Anthropic Claude, using ElevenLabs, Faster-Whisper, and Pygame.

Real-time voice AI agent with Amazon Nova Sonic and AWS Bedrock AgentCore. Features bidirectional streaming, React UI, WebSocket communication, and one-command deployment. Includes authentication, tool use, and local development mode.

Real-time bidirectional audio streaming for Expo and React Native. Record microphone input and play audio chunks with ultra-low latency using native AVAudioEngine. Built for voice AI applications, live audio processing, and real-time communication.

This project is a raw implementation of a low-latency, bidirectional real-time voice AI system built purely from foundational primitives (FastAPI, WebSockets, WebRTC VAD, AudioWorklets, streaming STT and token-level streaming TTS).

Real-time bidirectional voice AI conversations. Features WebSocket streaming, multi-agent support, and customizable voice profiles.

Voice-Bot is a premium AI voice assistant web application enabling real-time bidirectional conversations with Google Gemini AI. It uses WebSocket-based audio streaming for natural voice interaction with sub-500ms latency.

A complete voice conversation system built from scratch with real-time, bidirectional audio support in web browsers. Features low-latency audio streaming, WebSocket gateway, and robust interruption logic without relying on managed voice AI platforms.

a complete voice conversation system that runs in a web browser, supporting real-time, bidirectional audio. The primary task is to architect a foundational AI pipeline from scratch. This involves managing low-latency audio streaming, orchestrating backend services. without relying on pre-built voice AI platforms.

A lightweight C++ client library for real-time voice interaction with AI models running on edge servers. edge-vox enables bidirectional audio streaming and seamless integration with speech processing models like Whisper, LLaMA, and text-to-speech systems.