Self-host Duix.Avatar
v1.0.6Offline AI avatar cloning and video generation toolkit
docker-compose up -d- Stars
- 16 354
- RAM
- 32G or more
- Storage
- Free space greater than 100G
- Port
- 8383 (Video synthesis API (synthesis submit/query endpoints))
guiji2025/duix.avatarWhat is Duix.Avatar?
Duix.Avatar is an AI avatar (digital human) toolkit from Duix.com that clones a real person's appearance and voice from video and audio data, then renders talking-head videos entirely on your own machine. Text or recorded speech drives the avatar: automatic speech recognition and text-to-speech run in bundled fun-asr and fish-speech containers, and the synthesis service produces lip-synced video exposed over local HTTP APIs for model training, audio synthesis and video generation. The backend runs as three Docker services on an NVIDIA GPU, deployed with docker-compose from the /deploy directory on Windows 10 or Ubuntu 22.04 (a lite single-service compose variant is also provided); desktop clients ship as a Windows setup.exe installer and a Linux AppImage. Notable limits: all compute is local and needs an NVIDIA GPU with working drivers, the initial download is roughly 70GB of traffic, cloning works in non-real-time only (interactive conversation is the vendor's separate commercial API), and the DUIX.COM Community License requires a commercial license agreement above 1,000 monthly active users.
Duix.Avatar alternatives
OmniVoice-StudioThis self-hosted, open-source ElevenLabs alternative clones and designs voices, dubs video and builds audiobooks in 646 languages, fully local on desktop or Docker.CompareOmniVoice-Studio with Duix.Avatar
Fish SpeechSelf-hosted text-to-speech toolkit with voice cloning, multi-speaker scripts and inline emotion tags, plus a TTS API server and WebUI in CUDA or CPU Docker images.CompareFish Speech with Duix.Avatar
voice-chat-aiSelf-hosted open-source voice chat app for talking to AI characters in a web UI or terminal, mixing Ollama and cloud LLMs with Spark-TTS, Kokoro or ElevenLabs voices.Comparevoice-chat-ai with Duix.Avatar
Breeze TTS 2Self-hosted open-weight TTS model with voice cloning, design from descriptions, and voice direction. Streaming REST API and CLI for real-time synthesis on NVIDIA GPUs.CompareBreeze TTS 2 with Duix.Avatar
ComfyUISelf-hosted, open-source AI engine for visual creators with a node-based workflow system covering images, video, 3D, audio, and text.CompareComfyUI with Duix.Avatar
chatterbox-tts-apiChatterbox TTS API is a self-hosted, OpenAI-compatible TTS server with voice cloning and 22-language support. Runs via Docker or Python, integrates with Open WebUI.Comparechatterbox-tts-api with Duix.Avatar
ClipForgeSelf-hosted AI turns product images into short videos for Douyin, Kuaishou, Xiaohongshu, TikTok Shop — product-preserving visuals, AI scripts, voiceovers, subtitles.CompareClipForge with Duix.Avatar
Whisper-WebUISelf-hosted Gradio web UI that turns Whisper speech recognition into an easy subtitle generator: transcribe files, YouTube links or microphone.CompareWhisper-WebUI with Duix.Avatar
Should you self-host Duix.Avatar?
Duix.Avatar scores 68/100 on selfhost.directory (how scoring works) and ranks #22 of 123 in AI. It has 16 354 GitHub stars, +357 in the last 30 days.
There was no stable release in the past 12 months, so check the repository's recent activity before you rely on it.
Setup is advanced: plan for several services and some ongoing maintenance. An official Docker image is available. It is Other-licensed and written mostly in C.
Its closest alternative on selfhost.directory is openclaw (93/100, 394 567 stars). It scores higher overall, so compare the two before you pick. See Duix.Avatar vs openclaw side by side.
Features
- Digital human appearance cloning
- Fully offline operation
- Lip-sync video synthesis
- Model training from real video
- Multi-language Support
- Multiple model management
- Real-time interactive avatarReal-time interaction available via duix.com's commercial product only.
- REST API
- Text and voice-driven avatar video
- Voice Cloning
Duix.Avatar README
🚀🚀🚀 Duix Avatar — Truly open-source AI avatar toolkit for offline video generation and digital human cloning
🔗 Office website: www.duix.com
More in AI
Durable ActorsSelf-hosted runtime for stateful serverless actors with durable persistence and serialized execution, replacing Cloudflare Durable Objects.
DyadFree self-hosted open-source AI app builder running locally on desktop. Build web apps via natural language using OpenAI, Anthropic, Google or local models.
EchoSyncEchoSync is a self-hosted AI video generator turning text into multi-scene prototypes with narration, subtitles, and a digital anchor — all running locally via Docker.
elasticsearchOpen-source distributed search engine that handles structured, unstructured, and vector data for real-time search, observability, and AI workloads.
EmDashSelf-hosted open-source Astro CMS that runs on Node.js or Cloudflare, with a passkey admin, sandboxed plugins, REST API and a built-in MCP server.
emisaremisar gives AI agents declared infrastructure actions instead of shell. Policy gates actions, approvals for risky changes, tamper-evident audit trail.
EngraphisEngraphis is a self-hosted, open-source memory service for AI coding agents with durable scoped memory, local dashboard, MCP support, and bi-temporal history.
engrimEngrim is a self-hosted, open-source episodic memory engine that lets AI coding agents share project context across sessions via a local SQLite database.
EntrolyEntroly is a self-hosted, open-source tool that reduces AI token usage by picking relevant code evidence within a budget and providing auditable receipts.
envoy-ai-gatewayOpen-source AI gateway on Envoy Gateway. One OpenAI-compatible API for providers like OpenAI, Azure, Google Gemini, AWS Bedrock, Anthropic, and self-hosted vLLM.
EuConformSelf-hosted open-source toolkit for EU AI Act compliance: portable JSON artifacts for risk classification, bias testing, and AI BOM — offline and vendor-independent.
facetSelf-hosted photo manager using AI to score, cull, and organize your library locally with face detection and natural-language search.
0 comments