A podcast written by a room and checked by a transcript
Pitchers, a showrunner with a facts ledger, one segment at a time, and a transcript diff so the recording says what the script says. How After the Demo gets made.
Where it came from: How After the Demo, my two-host podcast, gets written, recorded and checked.
After the writers' room worked for short videos, I wanted the same thing for long form: a podcast about what happens after the AI demo, when someone actually has to run the thing. Two hosts, me and a co-host, recorded with AI voices including my own cloned one, built from real material.
Long form breaks in new ways. A 20-minute episode has to stay consistent across segments, and AI voices drift from the script. Two additions handle that.
The room, stretched out
- Three pitchers with different lenses (the listener, the operator, the provocateur) pitch angles from the week's material.
- A showrunner builds the rundown: cold open, intro, three to five segments of a few minutes each, outro. It also builds a facts ledger, every fact the episode is allowed to use, with its source.
- One segment at a time. The head writer drafts the dialogue, an editor critiques it with a red-line veto, and the writer revises. Then the segment is recorded and checked before the next one is written. That's what keeps voice and running jokes consistent.
The transcript check
The step I'd tell anyone to copy is this one. After recording, transcribe the audio locally and diff it against the script. Any line that drifted, dropped a word or mangled a product name gets re-recorded. The recorder also refuses to run if a banned word appears in the text, and a pronunciation map handles brand names.
Smaller rules came from listening: write numbers as words, keep the same voice settings and seed for the whole episode, level speech to around -16 LUFS, and keep music well under the voices. The synthetic co-host never claims a personal history. Sponsor mentions go in the show notes, not the conversation, and each episode mentions two products at most.
The result
Five episodes are live on soxoa.com/podcast and the major podcast apps. Every one went through the same stages, so I can trace any line back to the rundown and any fact back to its source.
Steal it
The skill is podcast-room in my John Lab plugin. It pairs with writers-room: use the same show bible for both.
/plugin marketplace add johnarndt/john-lab
/plugin install john-lab@john-lab/plugin marketplace add johnarndt/john-lab /plugin install john-lab@john-labDownload podcast-room.zipView on GitHub
View SKILL.md
---
name: podcast-room
description: "Use when a user wants to produce a two-host podcast episode with AI voices from real material. Triggers include 'make a podcast episode', 'turn this into a podcast', 'write a two-person segment', 'use my cloned voice', 'make a 20 minute episode', or 'set up a podcast feed'. Runs pitchers and a showrunner to build a rundown with a facts ledger, writes and edits each segment with a red-line veto, records with consistent voices, checks the recording against the script, and assembles a leveled episode."
---
# Podcast room
A sequel to `writers-room`, for long form. The trick is the same: separate roles, written stages, and a veto for truth. The new parts are continuity across segments and checking that the recording says what the script says.
## Show bible
`BIBLE.md`: hosts (name, voice, what each may claim about themselves; a synthetic co-host claims no personal history), audience, segment formats, red lines (no facts beyond the sources, nothing that identifies a customer, words that must never be spoken), product mention limit per episode, pronunciation map, and loudness target.
## Stages
1. **Pitch.** Three pitchers with different lenses (listener, operator, provocateur) each pitch angles from the sources: trending topics, launches, real work from the week.
2. **Rundown.** A showrunner picks the episode: cold open, intro, 3-5 segments of 3-5 minutes each, outro, and a **facts ledger** listing every fact the episode may use with its source.
3. **Segments, one at a time.** Head writer drafts `{"id","title","lines":[{"speaker","text"}]}` → editor critiques with a red-line veto → revise. Finish one segment (write, record, check) before starting the next, so voice and running jokes stay consistent.
4. **Record.** Same voice IDs and a fixed seed for the whole episode; apply the pronunciation map; abort if a banned word appears in the text.
5. **Check.** Transcribe the recording locally and diff it against the script. Re-record any line that drifted, dropped words or mispronounced a name.
6. **Assemble.** Cold open → jingle → intro → segments → outro, leveled to the loudness target (around -16 LUFS for speech), music well under the voices.
7. **Lint and publish prep.** Episode length, chapter markers, show notes with links (sponsor mentions go here, not in the dialogue), cover art, and a feed entry. The owner publishes.
## Rules
- Write numbers as words in scripts so voices read them naturally.
- Use only voices the owner has rights to. A cloned voice is the owner's own, used with their consent.
- Disclose synthetic voices wherever the platform or the law requires it.
- Never publish or submit the feed without the owner's approval.