A 5-workflow n8n pipeline that turns a single podcast or long-form video into a transcript, SEO blog post, social pack, newsletter, and viral clip timestamps — automatically. Self-hosted. One-time price. No subscriptions.
Most creators burn 4–8 hours per episode on manual repurposing — or rent 3+ tools that don't talk to each other.
"I publish my episode, then spend half a day turning it into a blog post, LinkedIn updates, X threads, a newsletter, and clips. I'm renting Castmagic, Opus Clip, Riverside Magic Clips, and a transcription service — $150–300/month once you're on the tiers that keep up with a weekly show — and they still don't connect to each other."
Import the JSON files into n8n. Plug in your OpenAI key. Pipe an episode in. Done.
OpenAI Whisper transcription with auto-generated chapter chunks, segment-level timestamps, and structured JSON output. Foundation for every other workflow.
audio_url → structured transcript + chapters
Produces a structured 1,500–2,500 word blog post with TL;DR, H2/H3 sections, pull quotes, and meta description. Optimized for search and skim-readers.
transcript → blog post (Markdown)
31+ pieces across 6 platforms: 3 X threads + 10 standalone tweets, 5 LinkedIn posts, 5 Instagram captions, 4 newsletter snippets, 4 YouTube Community posts. Each platform-tuned and ready to schedule.
transcript → 31+ posts across 6 platforms
Drafts a 400–600 word email newsletter with 3 subject-line options and preview text, plus structured show notes: summary, 5 key takeaways, timestamped topic markers, and Apple/Spotify descriptions.
transcript → email + show notes
Identifies high-engagement moments (hooks, controversies, surprising facts), ranks them, and outputs ready-to-cut clip timestamps with platform-tuned captions for Reels, Shorts, and TikTok.
transcript → ranked clip list w/ captions
Feed in a single 60-minute episode. Here's exactly what comes back — and what it costs.
Full timestamped transcript, auto-split into ~5-min chapters, with a segment array (start/end/text) and word count + duration metadata. Structured JSON the other four workflows read directly.
1,500–2,500 word article: SEO title, meta description, slug, primary + secondary keywords, H2/H3 structure, pull quotes, FAQ schema for featured snippets, and internal-link suggestions. HTML + Markdown.
31+ pieces: 3 X threads + 10 standalone tweets, 5 LinkedIn posts, 5 Instagram captions, 4 newsletter snippets, 4 YouTube Community posts. Each written to that platform's format.
400–600 word email with 3 subject-line options and preview text, plus show notes: summary, 5 takeaways, timestamped topic markers, resources list, and Apple/Spotify descriptions.
Top 10–15 clip candidates, each with start/end timestamps, a 1–10 viral score, hook/overlay text, a suggested caption, best-fit platforms, and a category tag.
All 5 workflow JSONs, a step-by-step setup guide (running in ~15 min), and an individual README for every workflow with customization notes. Lifetime access + future updates.
No subscriptions. No per-episode fees. No vendor lock-in.
Workflow 01 on its own. The foundation the rest of the pack builds on. No email required.
All 5 workflows. The full pipeline. First 50 buyers only.
Grab the transcript workflow free, no email required — download the JSON and setup guide directly.
Get the free workflowIf yours isn't here, email me and I'll add it.
No. The setup guide walks you through importing each JSON, connecting your OpenAI credential, and running your first episode. If you can install a Chrome extension, you can do this. Total setup: ~15 minutes.
OpenAI Whisper transcription is $0.006 per minute of audio, and the LLM calls (for blog/social/etc.) are typically $0.10–$0.50 per episode depending on length. A 60-minute episode costs about $0.50–$1.00 to run end-to-end. n8n cloud is $20/mo (or free if you self-host).
Yes — the pipeline accepts any audio file URL (RSS enclosures, S3, Dropbox direct links, Google Drive shared links). Works with Buzzsprout, Transistor, Spotify for Podcasters, Anchor, Megaphone, custom hosting, anything.
Absolutely — that's the whole point of self-hosted. Every prompt lives in a Code node you can edit, and each workflow's README shows you exactly where to change tone, language, or format.
Yes. Any video file URL with a reachable audio track works — the transcript workflow accepts the same audio URL input. For YouTube specifically, pull the audio to a file URL first (yt-dlp or any downloader), then feed that in.
Gumroad's standard policy: 30-day no-questions-asked refund. If a workflow doesn't import or run as advertised, just email and I'll refund.
Because (a) you'd pay $30–80/mo forever instead of $99 once, and (b) you'd be locked into my prompts and my infrastructure. This way you own the pipeline, customize anything, and run it on your own LLM credits.