Digital human

An agentic AI companion with a Live2D face: ElevenLabs STT/TTS over WebSocket, an OpenAI/xAI agentic loop, and shell, code, file and MCP tools behind an approval gate.

The JrsVision Unity package is a vision-based toolkit for scene understanding (action recognition for Virtual Humans, scene analysis etc.) .

Explore the 1000+ components inside a humanoid

An Attempt at simulating Human Body.

Bring your AI idol on air. An open-source studio for chat-driven generated speech and video.

Your own virtual idol, generated — not lip-synced. Chat makes her dance, eat, talk and change scenes, live. OSS, BYOK.

从个人定位、兴趣和参考图出发，逐步创建、固定并扩展可复用个人 IP 形象的 AI Agent Skill。

AI口播智能体,实体口播作弊神器

Retarget human motion from video to an Adobe Mixamo-rigged 3D character.

An experimental iPhone companion that gives voice models a face and personality, built with SwiftUI, WebGL, WebRTC and Blender.

Nvidia Audio2Face implementation inside blender

Provider-neutral Codex Skill for producing verified AI presenter videos from a script and an authorized presenter image.

Material-reference-driven photorealistic character casting skill for Codex

DeepSeek Harness 桌宠插件：元气鲸鱼娘陪你写代码 🐋

语音 AI 女友（Voice AI girlfriend for DeepSeek Harness）：Whisper 语音输入 + Qwen3-TTS 声音克隆 + 句子级流式朗读 + 数字人动画窗。插话/排队双模式，说话即打断。

A Clippy activity avatar for the DeepSeek Harness web client

Animated desktop companion for Google Antigravity 2.x with 193+ characters & real-time reasoning lifecycle sync. 首个专为 Antigravity 2.x 打造的桌面智能伴侣，开箱即用 193+ 款角色，全自动思考编码动作联动。桌宠 / 桌面宠物 / 二次元