transcript-intelligence
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseTranscript Intelligence
Transcript Intelligence
Overview
概述
Turn public video transcripts into useful research and content assets. This skill is for extracting signal from spoken social video: hooks, claims, stories, objections, examples, CTAs, and reusable content angles.
将公开视频的transcript转化为实用的研究和内容资产。此技能用于从社交视频的语音内容中提取关键信息:hooks、claims、故事、异议、示例、CTAs以及可复用的内容角度。
When to Use
使用场景
Use this skill when the user asks to:
- summarize a video, reel, short, TikTok, podcast clip, or social video
- analyze a creator's hooks or speaking style
- extract quotes, claims, examples, and CTAs from transcripts
- turn transcripts into social posts, scripts, newsletters, or content ideas
- compare what multiple creators say about a topic
当用户提出以下需求时,可使用此技能:
- 总结视频、Reel、短视频、TikTok、播客片段或社交视频
- 分析创作者的hooks或说话风格
- 从transcript中提取quotes、claims、示例和CTAs
- 将transcript转化为社交帖子、脚本、通讯或内容创意
- 对比多位创作者关于某一主题的观点
Transcript Sources
Transcript来源
| Platform | Endpoint |
|---|---|
| TikTok | |
| |
| YouTube | |
| |
| X/Twitter | |
| |
| Rumble | |
| Reddit video | |
If a detail endpoint already includes transcript text, use it. If transcript is unavailable, say so and fall back to title/caption/description only.
| 平台 | 端点 |
|---|---|
| TikTok | |
| |
| YouTube | |
| |
| X/Twitter | |
| |
| Rumble | |
| Reddit视频 | |
如果某个详情端点已包含transcript文本,则使用该端点。如果transcript不可用,请告知用户并仅使用标题/字幕/描述作为替代。
Workflow
工作流程
-
Collect URLs or discover videos
- If URLs are provided, fetch each transcript directly.
- If a creator/channel is provided, first fetch recent posts/videos, then choose relevant videos.
-
Extract transcript text
- Preserve timestamps if provided.
- Keep source URL with each transcript.
- Do not hallucinate missing captions.
-
Segment the transcript Break into:
- hook/opening
- setup/context
- main claim or lesson
- evidence/examples
- payoff
- CTA
-
Analyze the content Extract:
- exact hooks
- claims and contrarian takes
- stories
- frameworks
- objections addressed
- emotional language
- quotable lines
- content atoms that stand alone
-
Synthesize across multiple transcripts
- Cluster by topic and angle.
- Count recurring themes.
- Identify repeated hook formulas.
- Flag the strongest examples with citations.
-
收集URL或发现视频
- 如果提供了URL,直接获取每个transcript。
- 如果提供了创作者/频道,先获取近期帖子/视频,再选择相关视频。
-
提取transcript文本
- 如果提供时间戳则保留。
- 为每个transcript保留来源URL。
- 不要凭空生成缺失的字幕。
-
拆分transcript 拆分为:
- hook/开场
- 背景介绍/上下文
- 核心主张或要点
- 证据/示例
- 收益点
- CTA
-
分析内容 提取:
- 精准hooks
- 主张与反向观点
- 故事
- 框架
- 回应的异议
- 情绪化语言
- 可引用语句
- 可独立存在的content atoms
-
多transcript整合
- 按主题和角度聚类。
- 统计重复出现的主题。
- 识别重复使用的hook公式。
- 标记带有引用来源的最佳示例。
Output Format
输出格式
markdown
undefinedmarkdown
undefinedTranscript Intelligence Report
Transcript Intelligence Report
TL;DR
TL;DR
- Main themes:
- Strongest hooks:
- Best reusable ideas:
- Main themes:
- Strongest hooks:
- Best reusable ideas:
Transcript-by-Transcript Notes
Transcript-by-Transcript Notes
Video title or URL
Video title or URL
- Hook: "..."
- Core idea: ...
- Best quote: "..."
- CTA: ...
- Content atoms:
- ...
- Hook: "..."
- Core idea: ...
- Best quote: "..."
- CTA: ...
- Content atoms:
- ...
Patterns Across the Set
Patterns Across the Set
| Pattern | Evidence | Example URLs |
|---|
| Pattern | Evidence | Example URLs |
|---|
Hooks Swipe File
Hooks Swipe File
- "..."
- "..."
- "..."
- "..."
Repurposing Ideas
Repurposing Ideas
- LinkedIn post:
- X thread:
- Short-form script:
- Newsletter section:
undefined- LinkedIn post:
- X thread:
- Short-form script:
- Newsletter section:
undefinedCommon Pitfalls
常见误区
- Do not summarize from the title alone if the user asked for transcript analysis. Fetch transcripts first.
- Do not lose exact wording. Hooks and quotes are more valuable verbatim.
- Do not treat AI-generated transcript text as perfect. If wording seems garbled, mark it as approximate.
- Do not mix source attribution. Every quote should trace back to a URL.
- 如果用户要求进行transcript分析,不要仅通过标题进行总结。先获取transcript。
- 不要丢失原文措辞。Hooks和quotes保留原文更有价值。
- 不要将AI生成的transcript文本视为完美无缺。如果措辞看起来混乱,标记为近似内容。
- 不要混淆来源归属。每个quote都应追溯到对应的URL。