[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:ai-avatar-system":3},"\u003Cdiv align=\"center\">\u003Ch1>🎭 AvatarAI — Real-Time AI Avatar Platform\u003C\u002Fh1>\u003Cp>\u003Cstrong>Upload a photo · Clone a voice · Talk to any face in real time\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPunithVT\u002Fai-avatar-system\u002Fstargazers\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPunithVT\u002Fai-avatar-system?style=for-the-badge&color=7c3aed\" alt=\"Stars\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPunithVT\u002Fai-avatar-system\u002Fforks\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002FPunithVT\u002Fai-avatar-system?style=for-the-badge&color=3b82f6\" alt=\"Forks\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPunithVT\u002Fai-avatar-system\u002Fissues\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002FPunithVT\u002Fai-avatar-system?style=for-the-badge\" alt=\"Issues\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPunithVT\u002Fai-avatar-system\u002Fblob\u002FHEAD\u002FLICENSE\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg?style=for-the-badge\" alt=\"MIT License\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Cp>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FNext.js-14-black?logo=next.js&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FFastAPI-0.109-009688?logo=fastapi&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPython-3.10+-3776AB?logo=python&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTypeScript-5-3178C6?logo=typescript&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocker-Compose-2496ED?logo=docker&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCUDA-11.8-76B900?logo=nvidia&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPostgreSQL-15-336791?logo=postgresql&style=flat-square\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRedis-7-DC382D?logo=redis&style=flat-square\" \u002F>\n\u003C\u002Fp>\u003Cp>\n  \u003Ca href=\"#-quick-start\" rel=\"nofollow ugc noopener\">Quick Start\u003C\u002Fa> ·\n  \u003Ca href=\"#-features\" rel=\"nofollow ugc noopener\">Features\u003C\u002Fa> ·\n  \u003Ca href=\"#-architecture\" rel=\"nofollow ugc noopener\">Architecture\u003C\u002Fa> ·\n  \u003Ca href=\"#-gpu--aws-deployment\" rel=\"nofollow ugc noopener\">GPU \u002F AWS Deploy\u003C\u002Fa> ·\n  \u003Ca href=\"#-api-reference\" rel=\"nofollow ugc noopener\">API\u003C\u002Fa> ·\n  \u003Ca href=\"#-roadmap\" rel=\"nofollow ugc noopener\">Roadmap\u003C\u002Fa>\n\u003C\u002Fp>\u003Cblockquote>\n\u003Cp>\u003Cstrong>The most complete open-source AI avatar \u002F digital human system.\u003C\u002Fstrong>\nReal-time talking-head lip-sync · Zero-shot voice cloning · Multi-LLM · Runs 100% locally or on AWS.\u003C\u002Fp>\n\u003C\u002Fblockquote>\n\u003C\u002Fdiv>\u003Chr \u002F>\n\u003Ch2>🎬 What is AvatarAI?\u003C\u002Fh2>\n\u003Cp>AvatarAI is an open-source, production-ready platform for building \u003Cstrong>photorealistic AI avatar conversations\u003C\u002Fstrong>. Upload any face photo, clone a voice from a 5-second audio clip, and have a real-time conversation — with \u003Cstrong>lip-sync video generated on every single response\u003C\u002Fstrong>.\u003C\u002Fp>\n\u003Cpre>\u003Ccode>[mic] → Whisper STT → Claude \u002F GPT \u002F Ollama (streaming) → Chatterbox TTS → MuseTalk lip-sync → [video]\n                              < 2–4 s to first video chunk on AWS GPU >\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>\u003Cstrong>What makes AvatarAI different:\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Cul>\n\u003Cli>🎤 \u003Cstrong>Zero-shot voice cloning\u003C\u002Fstrong> — 10 seconds of audio is all you need (Chatterbox Multilingual)\u003C\u002Fli>\n\u003Cli>🎭 \u003Cstrong>Any face, any language\u003C\u002Fstrong> — upload a JPEG, pick from 23 languages, start talking\u003C\u002Fli>\n\u003Cli>⚡ \u003Cstrong>Token-streaming pipeline\u003C\u002Fstrong> — the LLM streams live tokens while TTS + lip-sync run per sentence; the first video chunk plays before the model finishes its reply\u003C\u002Fli>\n\u003Cli>✋ \u003Cstrong>Barge-in\u003C\u002Fstrong> — speak (or hit stop) mid-reply and the avatar yields instantly, like a real conversation\u003C\u002Fli>\n\u003Cli>🔒 \u003Cstrong>100% local mode\u003C\u002Fstrong> — local storage, local Whisper, local LLM via Ollama: nothing leaves your machine\u003C\u002Fli>\n\u003Cli>🔌 \u003Cstrong>Multi-LLM\u003C\u002Fstrong> — Claude (with prompt caching), GPT-4o, or any local model via Ollama \u002F vLLM \u002F LM Studio\u003C\u002Fli>\n\u003Cli>🚀 \u003Cstrong>AWS GPU deployment\u003C\u002Fstrong> — one-command deploy to \u003Ccode>g5.xlarge\u003C\u002Fcode> for true real-time (~30 FPS)\u003C\u002Fli>\n\u003Cli>🏗️ \u003Cstrong>Production-grade\u003C\u002Fstrong> — JWT + httpOnly-cookie auth, per-user rate limiting, Postgres + Alembic, S3\u002FCloudFront, Prometheus, CI, a real test suite — the only project in this niche you can ship as a product, not just a demo\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Chr \u002F>\n\u003Ch2>⚔️ How AvatarAI compares\u003C\u002Fh2>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>\u003C\u002Fth>\n\u003Cth>\u003Cstrong>AvatarAI\u003C\u002Fstrong>\u003C\u002Fth>\n\u003Cth>Duix-Avatar\u003C\u002Fth>\n\u003Cth>Linly-Talker\u003C\u002Fth>\n\u003Cth>AIAvatarKit\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>Real-time conversation\u003C\u002Ftd>\n\u003Ctd>✅ WebSocket streaming\u003C\u002Ftd>\n\u003Ctd>❌ offline video gen\u003C\u002Ftd>\n\u003Ctd>✅ (Gradio \u002F WebRTC spin-off)\u003C\u002Ftd>\n\u003Ctd>✅\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>Lip-sync video\u003C\u002Ftd>\n\u003Ctd>✅ MuseTalk V1.5\u003C\u002Ftd>\n\u003Ctd>✅ proprietary models\u003C\u002Ftd>\n\u003Ctd>✅ multiple engines\u003C\u002Ftd>\n\u003Ctd>❌ (drives external avatars)\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>Voice cloning\u003C\u002Ftd>\n\u003Ctd>✅ 10 s, 23 languages\u003C\u002Ftd>\n\u003Ctd>✅\u003C\u002Ftd>\n\u003Ctd>✅\u003C\u002Ftd>\n\u003Ctd>❌\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>Barge-in \u002F interr\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n",1788219545121]