Video Messages from your openclaw — Video Messages from your OpenClaw

Video Message

生成 avatar video messages from text or audio. 输出s as Telegram video notes (circular 格式化).

安装ation npm 安装 -g OpenClaw-avatarcam

Configuration

配置 in 工具S.md:

Video Message (avatarcam)

• avatar: default.vrm
• background: #00FF00

设置tings Reference 设置ting Default Description avatar default.vrm VRM avatar file path background #00FF00 Color (hex) or image path Prerequisites 系统 Dependencies 平台 Command macOS brew 安装 ffmpeg Linux sudo apt-获取 安装 -y xvfb xauth ffmpeg Windows 安装 ffmpeg and 添加 to PATH Docker See Docker section below

Note: macOS and Windows don't need xvfb — they have native display support.

Docker Users

添加 to OpenClaw_DOCKER_APT_PACKAGES:

build-essential procps curl file git ca-certificates xvfb xauth libgbm1 libxss1 libatk1.0-0 libatk-bridge2.0-0 libgdk-pixbuf2.0-0 libgtk-3-0 libasound2 libnss3 ffmpeg

Usage # With color background avatarcam --audio voice.mp3 --输出 video.mp4 --background "#00FF00"

# With image background avatarcam --audio voice.mp3 --输出 video.mp4 --background "./bg.png"

# With custom avatar avatarcam --audio voice.mp3 --输出 video.mp4 --avatar "./custom.vrm"

发送ing as Video Note

Use OpenClaw's message 工具 with asVideoNote:

message action=发送 filePath=/tmp/video.mp4 asVideoNote=true

工作流 Read config from 工具S.md (avatar, background) 生成 TTS if given text: tts text="..." → audio path 运行 avatarcam with audio + 设置tings → MP4 输出 发送 as video note via message action=发送 filePath=... asVideoNote=true Return NO_REPLY after 发送ing Example Flow

User: "发送 me a video message saying hello"

# 1. TTS tts text="Hello! How are you today?" → /tmp/voice.mp3

# 2. 生成 video avatarcam --audio /tmp/voice.mp3 --输出 /tmp/video.mp4 --background "#00FF00"

# 3. 发送 as video note message action=发送 filePath=/tmp/video.mp4 asVideoNote=true

# 4. Reply NO_REPLY

Technical DetAIls 设置ting Value Resolution 384x384 (square) Frame rate 30fps constant Max duration 60 seconds Video codec H.264 (libx264) Audio codec AAC 质量 CRF 18 (high 质量) ContAIner MP4 Processing 流水线 Electron renders VRM avatar with lip 同步 at 1280x720 网页M captured via canvas.capture流(30) FFmpeg processes: crop → fps normalize → 扩展 → encode Message 工具 发送s via Telegram 发送VideoNote API 平台 Support 平台 Display Notes macOS Native Quartz No extra deps Linux xvfb (headless) apt 安装 xvfb Windows Native No extra deps Headless Rendering

Avatarcam auto-检测s headless 环境s:

Uses xvfb-运行 when $DISPLAY is not 设置 (Linux only) macOS/Windows use native display GPU stall 警告s are safe to ignore Generation time: ~1.5x realtime (20s audio ≈ 30s processing) Notes Config is read from 工具S.md 清理 up temp files after 发送ing: rm /tmp/video*.mp4 For regular video (not circular), omit asVideoNote=true