Video Transcript Optimizer For SEO Repurposing

Overview

Video Transcript Optimizer for SEO & Repurposing transforms raw video transcripts into a complete suite of repurposed content assets in minutes. This skill solves the creator's dilemma: "I have 50 hours of video content but can't efficiently distribute it across blogs, social media, email, and search engines."

The skill intelligently extracts:

• SEO-optimized video descriptions (YouTube, Vimeo, WordPress)
• Chapter timestamps with auto-generated summaries
• Social media clips with burned-in captions (TikTok, Instagram Reels, YouTube Shorts format)
• Blog post outlines with H2/H3 structure ready for expansion
• Quotable moments formatted as shareable graphics
• Speaker identification & attribution for multi-speaker content
• Multi-language support (transcription and translation)

Why it matters: Content creators and agencies spend 15+ hours manually repurposing a single hour of video. This skill cuts that to 5 minutes per video, multiplying content ROI across WordPress blogs, LinkedIn, Twitter/X, email newsletters, and YouTube channels.

Integration ecosystem: Works with YouTube Data API, Google Cloud Speech-to-Text, OpenAI GPT-4, WordPress REST API, Slack webhooks, and standard video formats.

Quick Start

Try these prompts immediately to see the skill in action:

Example 1: Basic Transcript Processing

Process this video transcript for repurposing:

[00:00] Host: "Today we're talking about AI in marketing."
[00:15] Guest: "AI is transforming how we work..."
[00:45] Host: "What's the biggest challenge?"
[01:20] Guest: "Finding quality training data..."
[02:10] Host: "Any recommendations?"
[02:35] Guest: "Start with open-source models..."
[03:00] Host: "Thanks for being here."

Extract: SEO description, 3 chapter breaks, 2 pull quotes, and a blog outline.

Example 2: Multi-Language Content

I have a Spanish-language webinar transcript (45 minutes). 
- Extract SEO description in Spanish AND English
- Identify speaker names automatically
- Create 4 social clips (Instagram Reels format: 9:16 vertical)
- Generate blog outline in both languages
- Find top 3 quotable moments

Example 3: Timestamped Editing with Segments

My transcript has speaker labels [SPEAKER: Name]. 
Extract only segments from "Product Manager" speaker.
Create a 5-minute highlight reel outline with timestamps.
Format for YouTube Shorts (auto-add captions).
Generate LinkedIn post from these highlights.

Example 4: Batch Processing with Metadata

I have 10 podcast transcripts (CSVs with columns: Title, Transcript, Duration, Guest).
For each:
- Generate YouTube SEO description (160 chars)
- Extract top 2 quotable moments as image templates
- Create chapter markers (target 5-7 chapters)
- Output as WordPress blog outline
- Save metadata to JSON for bulk upload

Capabilities

1. SEO Description Generation

Automatically crafts YouTube, Vimeo, and WordPress-optimized descriptions (160-500 characters) that:

• Include primary keyword in first sentence
• Incorporate 3-5 long-tail keywords naturally
• Add CTA (link to blog, signup, product)
• Include timestamp chapters for accessibility
• Follow YouTube's best practices for discoverability

Example output:

Learn how AI transforms marketing strategy without coding. 
Guest expert shares proven frameworks for automating campaigns, 
finding training data, and scaling with open-source models. 
Perfect for marketers, product managers, and startup founders. 
[00:00 Intro] [00:45 AI Challenges] [02:35 Recommendations]
→ Read full guide: [link]

2. Intelligent Chapter Extraction

Analyzes transcript content to identify natural breakpoints and creates:

• Auto-generated chapter titles (3-8 per hour of video)
• Precise timestamp markers (within 5-10 seconds)
• Chapter summaries (1-2 sentences each)
• SEO-friendly slug format for YouTube chapters

Usage:

Extract chapters from this 28-minute sales webinar transcript.
Target: 6-8 chapters, each 3-5 minutes long.
Naming convention: "Section: Specific Topic"
Include chapter descriptions for YouTube SEO.

3. Social Media Clip Generation

Extracts and formats video clips for platform-specific distributions:

• TikTok/Reels: 9:16 vertical format, auto-caption placement
• YouTube Shorts: 1080×1920, burn-in captions
• Twitter/X: Tweet-sized quotes (280 chars) with video timestamp
• LinkedIn: Professional clips with context headers
• Outputs as timestamp ranges + caption JSON for automated video editing

Example:

{
  "clips": [
    {
      "start": "00:45",
      "end": "01:20",
      "platform": "reels",
      "caption": "AI isn't replacing marketers. It's replacing marketers who don't learn AI.",
      "cta": "Watch the full interview →"
    }
  ]
}

4. Blog Post Outline Generation

Transforms transcripts into WordPress-ready blog structures:

• H2 headers from chapter breaks
• H3 subheadings from key discussion points
• Suggested image breakpoints (every 2-3 min of video)
• Internal linking recommendations
• Meta description suggestions
• Estimated read time and word count targets

Output format: Markdown or WordPress XML ready for bulk import

5. Quote Graphics & Pullquote Extraction

Identifies memorable, shareable moments and outputs:

• Top 5-10 quotable moments ranked by engagement potential
• Text formatting with speaker attribution
• Suggested visual templates (Canva-compatible JSON)
• Image dimensions for Instagram (1080×1350), Pinterest (1000×1500), Twitter (1200×628)
• Hashtag recommendations per quote

6. Speaker Identification & Attribution

• Auto-detects speaker names from labeled transcripts
• Resolves speaker consistency (handles "the host" → "John Smith")
• Extracts speaker bios/roles from context
• Generates speaker cards for blog headers
• Creates podcast guest profiles for WordPress custom post types

7. Multi-Language Support

• Transcription: Converts audio transcripts into Spanish, French, German, Mandarin, Japanese
• Automatic translation: Creates parallel transcripts and bilingual SEO descriptions
• Cultural adaptation: Adjusts tone and references for target markets
• Subtitle generation: SRT/VTT format for captions in multiple languages

Configuration

Environment Variables

# Required
OPENAI_API_KEY=sk-... # GPT-4 for intelligent extraction
GOOGLE_CLOUD_API_KEY=... # Speech-to-Text & Translation API

# Optional
WORDPRESS_REST_URL=https://yourblog.com/wp-json/wp/v2
WORDPRESS_REST_USER=...
WORDPRESS_REST_PASS=...
SLACK_WEBHOOK_URL=https://hooks.slack.com/services/...

Configuration Options

{
  "extraction": {
    "targetChapters": 6,
    "minChapterDuration": 180,
    "includeTimestamps": true,
    "speakerLabeling": true
  },
  "seo": {
    "keywordDensity": "1-2%",
    "descriptionLength": 160,
    "includeCTA": true
  },
  "socialClips": {
    "formats": ["reels", "shorts", "tiktok"],
    "captionBurnIn": true,
    "minimumClipLength": 15
  },
  "languages": ["en", "es", "fr"],
  "output": {
    "format": "json",
    "includeImages": true,
    "wordpressExport": true
  }
}

Setup Instructions

1. Obtain API keys:

   • OpenAI: https://platform.openai.com/api-keys
   • Google Cloud: https://console.cloud.google.com

2. Install dependencies:

   pip install openai google-cloud-speech google-cloud-translate
   

3. Configure environment:

   export OPENAI_API_KEY="your-key"
   export GOOGLE_CLOUD_API_KEY="your-key"
   

4. Test the skill:

   openclaw run video-transcript-optimizer-for-seo-repurposing \
     --input "sample-transcript.txt" \
     --config config.json
   

Example Outputs

Sample 1: Complete Repurposing Package

Input: 32-minute product launch webinar transcript

Outputs generated:

📄 OUTPUTS/
├── youtube_description.txt
│   "Discover how our new AI platform transforms customer support. 
│    Watch Co-founder Sarah explain the 3-part framework, live demos, 
│    and roadmap. [00:00 Intro] [05:20 The Problem] [12:45 Demo] 
│    [24:10 Pricing & Roadmap] [30:00 Q&A]"
│
├── chapters.json
│   {
│     "chapters": [
│       {"start": "00:00", "title": "Introduction", "description": "Welcome & agenda"},
│       {"start": "05:20", "title": "Problem Statement", "description": "Why legacy systems fail"},
│       {"start": "12:45", "title": "Live Product Demo", "description": "End-to-end walkthrough"},
│       {"start": "24:10", "title": "Pricing & Roadmap", "description": "Plans and 2024 features"}
│     ]
│   }
│
├── social_clips.json
│   {
│     "clips": [
│       {
│         "title": "The Biggest Mistake",
│         "start": "07:15", "end": "08:45",
│         "platforms": ["reels", "shorts", "tiktok"],
│         "caption": "90% of teams are still using spreadsheets for customer support. It's costing them $200K+ annually.",
│         "imageTemplate": "quote-dark-blue"
│       },
│       {
│         "title": "Live Demo Moment",
│         "start": "14:30", "end": "16:50",
│         "platforms": ["youtube", "linkedin"],
│         "caption": "Watch the AI respond to 50 support tickets in seconds",
│         "cta": "See demo →"
│       }
│     ]
│   }
│
├── blog_outline.md
│   # How Our AI Platform Transforms Customer Support
│   ## The Problem: Legacy Systems Are Broken
│   ### Why Spreadsheets Don't Scale
│   ### Real Cost of Manual Support
│   ## The Solution: AI-Powered Triage
│   ### Intelligent Ticket Routing
│   ### Automated Response Generation
│   ## Live Demo: See It In Action
│   ## Pricing & Roadmap
│   ### 3-Tier Pricing Model
│   ### 2024 Feature Roadmap
│   ## Getting Started
│
├── quotes.json
│   {
│     "quotes": [
│       {
│         "text": "90% of teams are still using spreadsheets for customer support.",
│         "speaker": "Sarah Chen",
│         "timestamp": "07:20",
│         "engagement_score": 0.94,
│         "platforms": ["linkedin", "twitter", "pinterest"],
│         "imageTemplates": ["minimal-light", "bold-dark", "gradient-modern"]
│       },
│       {
│         "text": "We've reduced response time from 24 hours to 2 minutes.",
│         "speaker": "Sarah Chen",
│         "timestamp": "25:45",
│         "engagement_score": 0.89,
│         "platforms": ["all"]
│       }
│     ]
│   }
│
└── wordpress_import.xml
    <item>
      <title>How Our AI Platform Transforms Customer Support</title>
      <description>Learn the 3-part framework...</description>
      <content:encoded><![CDATA[
        <h2>The Problem: Legacy Systems Are Broken</h2>
        <p>[Embed video with timestamp: 5:20]</p>
        <p>90% of teams are still using spreadsheets...</p>
      ]]></content:encoded>
    </item>

Sample 2: Multi-Language Output

Input: Spanish webinar transcript (45 min)

{
  "en": {
    "seoDescription": "Learn advanced SEO strategies from industry expert...",
    "chapters": [...],
    "blogOutline": "..."
  },
  "es": {
    "seoDescription": "Aprende estrategias de SEO avanzadas del experto...",
    "chapters": [...],
    "blogOutline": "..."
  },
  "socialClips": {
    "clips": [
      {
        "caption_en": "The #1 SEO mistake most teams make...",
        "caption_es": "El error #1 de SEO que cometen la mayoría...",
        "timestamp": "12:45-14:20"
      }
    ]
  }
}

Tips & Best Practices

1. Transcript Quality = Output Quality

• Best: Professional transcription with speaker labels, minimal errors
• Good: Automated captions with 95%+ accuracy
• Acceptable: Raw transcripts with 5-10% errors (skill corrects common terms)
• Poor: Highly corrupted or low-quality audio transcripts

Pro tip: Use Otter.ai, Rev, or Google Cloud Speech-to-Text for best results before importing.

2. Optimize Chapter Structure

• Aim for 5-8 chapters per hour of video content
• Use semantic breaks (topic changes) rather than time intervals
• Name chapters with action words: "Discover," "Learn," "Master," "Avoid"
• Keep chapters 3-5 minutes for YouTube Shorts repurposing

3. Leverage Speaker Labels for Authority

• Always include speaker names and titles in transcript
• Format: [SPEAKER: John Smith, CEO] or [HOST:] / [GUEST:]
• Enables automatic speaker cards, LinkedIn attributions, and podcast guest profiles
• Improves engagement metrics when audiences know who's speaking

4. Plan Social Clip Strategy

• Extract 2-3 clips per 10 minutes of video for maximum coverage
• Prioritize contrarian takes, data points, and tips for high engagement
• Use clips to drive traffic back to long-form content (blog, YouTube)
• Schedule clips across 3-7 days to maximize reach without saturation

5. SEO-First Blog Outlines

• Extract outline before writing blog post
• Use generated H2/H3 structure as skeleton, add 30-50% new content
• Include video embeds at chapter breaks for dwell time and backlinks
• Add internal links to related posts (suggested by skill)

6. Multi-Language = Expanded Reach

• Translate to 2-3 high-value languages based on audience (Spanish, Portuguese, French most ROI)
• Use bilingual SEO descriptions to rank in both language searches
• Schedule translated content with 2-week stagger to avoid cannibalization

7. Automate Downstream Publishing

• Export WordPress XML directly to bulk upload 10+ blog posts weekly
• Use Zapier/Make integration with