跳到正文
TakoAPI
返回技能列表

veed-ugc

由 pauldelavallaz 提供

Coding Agents & IDEs
818下载量由 clawskills.sh 提供
5星标
0浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

Generate UGC-style promotional videos with AI lip-sync.

描述

该技能的功能

Converts a static person-with-product image into a lip-synced promotional video using a plain dialogue script. Powered by ComfyDeploy's Veed-UGC workflow and ElevenLabs for voice synthesis. Output is an MP4 with animated speech and natural head movements.

Skips the need for live video shoots by turning any frontal photo into a speaking promotional clip.

何时使用

  • Creating UGC-style ads from product photos
  • Scaling promotional video content without on-camera talent
  • Generating multilingual promos by swapping scripts
  • Animating Morpheus or Ad-Ready outputs into speaking videos
  • Producing quick social ad variations from a single image

示例工作流

输入

User asks: create a UGC promo video from this product shot with the script 'Mirá, yo antes no usaba esto pero ahora no puedo vivir sin él.'

智能体
  1. 1

    Accepts the image URL or local file path

  2. 2

    Validates the script contains pure dialogue with no annotations or stage directions

  3. 3

    Calls the ComfyDeploy API with the image, script, and voice_id

  4. 4

    Polls for job completion (processing takes 2-5 minutes)

  5. 5

    Downloads and saves the resulting MP4 file

输出

ugc-promo.mp4 — a lip-synced video of the person speaking the provided script

前置要求

  • ComfyDeploy API key
  • ElevenLabs voice ID (default provided, custom voices require ElevenLabs account)

安装

clawhub install pauldelavallaz/veed-ugc

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接