360 Secure SkillHub
360 安全技能中心

Script to Video

面向已有成稿脚本的用户：校验脚本文本后调用 Revid 公共渲染接口，按 9:16 等比例生成带配音、自动匹配素材画面、字幕与音乐的成片，并可轮询渲染状态；同时给出脚本过长、配音发音、素材不匹配和音乐盖过人声等失败场景的处理建议。

api00
api00
热度 33.42
最近更新 2026-05-11 17:45:58
分类：创作依赖：代码执行 / 第三方 API应用场景：视频编辑 / 社交媒体

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add revid-script-to-video
Metadata
name
revid-script-to-video
description
Turn an already-written script into a video with voiceover, auto-cut stock visuals, and captions. Use when the user has the words and wants Revid to handle production.
metadata
{"openclaw":{"requires":{"config":["REVID_API_KEY"]}}}

Script → video

The script-to-video workflow is the lowest-friction path: you bring the words, Revid brings the visuals + voice + captions + edit.

When to use this skill

  • The user pastes a finished script (or generates one in-conversation).
  • They want full creative control over the words.
  • They are happy with stock visuals (otherwise see revid-script-with-custom-media).
  • For an idea-to-video flow, use revid-prompt-to-video.

Inputs

Field Required Notes
text yes The script. Use line breaks for scene boundaries.
aspectRatio no Default 9:16
voiceId no Pick to match the script tone
targetDuration no Auto-derived from script length if omitted

Step-by-step

  1. Validate text is non-trivial (>30 words) and within practical limits (~1500 words for a 5-min video).
  2. POST /render.
  3. Poll /status.

API call template

POST /api/public/v3/render
Host: www.revid.ai
Content-Type: application/json
key: $REVID_API_KEY

{
  "workflow": "script-to-video",
  "source":   { "text": "{SCRIPT}" },
  "aspectRatio": "9:16",
  "voice":    { "enabled": true, "voiceId": "aria-en-us", "stability": 0.6, "speed": 1.0, "language": "en-US" },
  "captions": { "enabled": true, "position": "middle", "autoCrop": true },
  "music":    { "enabled": true, "syncWith": "beats" },
  "media": {
    "type": "stock-video",
    "density": "medium",
    "animation": "soft",
    "quality": "pro",
    "imageModel": "good",
    "videoModel": "pro"
  },
  "options": {
    "summarizationPreference": "no-summarization",
    "soundEffects": true,
    "hasToGenerateCover": true
  },
  "render": { "resolution": "1080p", "frameRate": 30 }
}

summarizationPreference: "no-summarization" — the user wrote the script for a reason. Don't paraphrase it.

Examples

Failure modes

Symptom Fix
Script too long → exceeds context Either split into multiple /render calls (one per chapter) or set summarizationPreference: "summarize".
Voice mispronounces brand names Inline phonetic spelling in the script ("Revid (rev-id)").
Visuals don't match niche topic Pre-author a few key shots and switch to revid-script-with-custom-media.
Music drowns the voice Lower music duck — currently no direct knob; switch music.enabled: false and add ambient sound effects via options.soundEffects: true.

See also