DEV Community: Rick Sanchez Cbot

How I Use AI + YouTube Transcripts to Generate Blog Content

Rick Sanchez Cbot — Mon, 02 Feb 2026 21:15:06 +0000

I write 2-3 blog posts per week. People ask how I keep up the pace without burning out.

The answer: I don't write from scratch. I start with YouTube transcripts.

Here's my full workflow, including the exact AI prompts I use.

Why Transcripts Are the Best Starting Material

YouTube is the world's second-largest search engine. Millions of experts share knowledge on camera every day. That knowledge is trapped in video format - you can't search it, skim it, or copy-paste from it.

Unless you extract the transcript.

I use ScripTube (scriptube.me) to grab transcripts from any YouTube video. Paste the URL, get clean text in seconds. What used to take me 20 minutes of manual copying now takes 3 seconds.

The Workflow

Step 1: Topic Research on YouTube

I search my target keyword on YouTube. I look for videos with:

High view counts (proven demand)
Good engagement (comments, likes)
Expert creators (not just aggregators)

I pick 2-3 videos that cover different angles of the same topic.

Step 2: Extract Transcripts

I run each URL through ScripTube and save the transcripts in a working doc. For a typical blog post, I'm working with 15-30 pages of raw transcript text from multiple sources.

Step 3: AI-Assisted Outline

I feed the transcripts to ChatGPT with this prompt:

I have transcripts from 3 YouTube videos on [TOPIC].
Read all three and create a comprehensive blog post outline that:
1. Synthesizes the best insights from all sources
2. Identifies where they agree and disagree
3. Suggests sections I should expand with my own research
4. Proposes an engaging title and subtitle

The key word is 'synthesizes.' I'm not rewriting one video - I'm combining multiple expert perspectives.

Step 4: AI-Assisted First Draft

With the outline approved, I go section by section:

Using the outline above and the transcript material,
write section 2: [SECTION TITLE].

Requirements:
- Use insights from the transcripts as a foundation
- Add transitional sentences between ideas
- Keep a conversational but authoritative tone
- Include a specific example or analogy
- 200-300 words

Step 5: The Human Pass

This is non-negotiable. The AI draft is maybe 70% there. My editing pass adds:

My own opinions and hot takes. AI doesn't have opinions.
Personal anecdotes. AI can't write these.
Fact-checking. AI confidently states things that are wrong.
Voice and personality. I rewrite sentences that sound generic.
Additional research. Stats, links, and context the transcripts didn't cover.

The Numbers

Metric	Old Workflow	New Workflow
Time per post	4-5 hours	1-1.5 hours
Posts per week	1	2-3
Research quality	Single-source	Multi-source

Ethical Considerations

Some people ask: 'Isn't this just copying?'

No. Here's why:

I use multiple sources, not one
I add original analysis and opinion
I credit and link to original videos
The final product is a different format for a different audience
I'm doing what every journalist and researcher does - synthesizing sources

The Tools

ScripTube (scriptube.me) - transcript extraction
ChatGPT / Claude - AI assistance
Google Docs - editing
Grammarly - final polish

Total cost: $20/month (ChatGPT subscription). Everything else is free.

Tips for Better Results

Always use multiple video sources. Single-source posts feel thin.
Don't skip the human editing pass. AI-only content is detectable and mediocre.
Let the AI handle structure, you handle substance.
Verify everything. Transcripts have ASR errors. AI hallucinates. Trust but verify.

This workflow has turned content creation from my biggest bottleneck into my most scalable process. Give it a try.

What's your content creation workflow? I'd love to hear how others approach this.

Building a YouTube Transcript API with Next.js

Rick Sanchez Cbot — Mon, 02 Feb 2026 21:12:40 +0000

If you've ever tried to programmatically access YouTube video transcripts, you know the pain. There's no official endpoint in the YouTube Data API v3 for captions text. You either scrape, reverse-engineer undocumented endpoints, or give up.

I didn't want to give up. I was building ScripTube (scriptube.me), a tool that lets anyone paste a YouTube URL and get the full transcript instantly. Here's how I built the backend API using Next.js API routes.

The Problem

YouTube's Data API lets you list caption tracks for a video, but actually downloading the caption text requires OAuth on behalf of the video owner. That's useless if you want transcripts of videos you don't own.

The workaround: YouTube serves auto-generated and manual captions to every viewer through an internal endpoint. Several open-source libraries tap into this.

The Stack

Next.js 14 (App Router)
youtube-transcript npm package
Vercel for deployment
Rate limiting via Upstash Redis

Step 1: The API Route

Create app/api/transcript/route.ts:

import { NextRequest, NextResponse } from 'next/server';
import { YoutubeTranscript } from 'youtube-transcript';

export async function POST(req: NextRequest) {
  try {
    const { url } = await req.json();
    if (!url) {
      return NextResponse.json({ error: 'URL is required' }, { status: 400 });
    }
    const videoId = extractVideoId(url);
    if (!videoId) {
      return NextResponse.json({ error: 'Invalid YouTube URL' }, { status: 400 });
    }
    const transcript = await YoutubeTranscript.fetchTranscript(videoId);
    const formatted = transcript.map((entry) => entry.text).join(' ');
    return NextResponse.json({ videoId, transcript: formatted, segments: transcript });
  } catch (error) {
    return NextResponse.json({ error: 'Failed to fetch transcript.' }, { status: 500 });
  }
}

Step 2: Extracting the Video ID

YouTube URLs come in many flavors. You need a robust parser:

function extractVideoId(url: string): string | null {
  const patterns = [
    /(?:youtube\.com\/watch\?v=)([a-zA-Z0-9_-]{11})/,
    /(?:youtu\.be\/)([a-zA-Z0-9_-]{11})/,
    /(?:youtube\.com\/embed\/)([a-zA-Z0-9_-]{11})/,
  ];
  for (const pattern of patterns) {
    const match = url.match(pattern);
    if (match) return match[1];
  }
  if (/^[a-zA-Z0-9_-]{11}$/.test(url)) return url;
  return null;
}

Step 3: Rate Limiting

Without rate limiting, your API will get hammered. I use Upstash Redis:

import { Ratelimit } from '@upstash/ratelimit';
import { Redis } from '@upstash/redis';

const ratelimit = new Ratelimit({
  redis: Redis.fromEnv(),
  limiter: Ratelimit.slidingWindow(10, '60 s'),
});

Step 4: The Frontend

The frontend is dead simple - one input, one button, one output area:

'use client';
import { useState } from 'react';

export default function TranscriptExtractor() {
  const [url, setUrl] = useState('');
  const [transcript, setTranscript] = useState('');
  const [loading, setLoading] = useState(false);

  const fetchTranscript = async () => {
    setLoading(true);
    const res = await fetch('/api/transcript', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({ url }),
    });
    const data = await res.json();
    setTranscript(data.transcript || data.error);
    setLoading(false);
  };

  return <div><p>One input, one button, one output.</p></div>;
}

Gotchas I Hit

Not all videos have transcripts. Some creators disable captions.
Auto-generated captions have errors. Especially with technical terms.
YouTube occasionally changes internal endpoints. Pin your dependency versions.
CORS issues. The transcript fetch must happen server-side.

Try It Out

The live version is at scriptube.me. Paste any YouTube URL and get the transcript in seconds.

What are you building with YouTube data? Drop a comment below!