On this page
- Why audio wins the in-between moments
- The format that actually gets listened to
- The 15-minute workflow
- 1. Set the strategic guardrails (3 min)
- 2. Curate the narrative flow (2 min)
- 3. Edit the dialogue (3 min)
- 4. Direct the voices and the emotion (4 min)
- 5. Translate, if you have international readers (optional)
- 6. Generate, export, distribute (3 min)
- What this unlocks
If you spent ten hours writing your last Medium article, you probably also watched the read-through rate quietly tell you that 80% of the people who clicked never finished it. That's not a writing problem. It's an attention problem — and it's the single biggest tax on modern thought leadership.
The fix isn't more text. It's a second channel for the same idea: audio.
Why audio wins the in-between moments
A 2,000-word article competes for 100% of the reader's visual attention. After eight hours of video calls and Slack, that's a cognitive ask most professionals quietly decline — they bookmark for "later," and later doesn't come.
Audio is different. It rides shotgun during the commute, the gym session, the dog walk. It uses the brain's default-mode network, which lets people absorb complex ideas while doing something physical that doesn't compete for the eyes. That's where the missing 80% of your readers actually live.
We call this completion arbitrage: you don't create more content, you simply meet the same audience in a moment when they can finish it.
The format that actually gets listened to
Nobody finishes a robotic monologue of a Medium article. The format that works is the one human ears are evolutionarily wired for: two people talking.
- The peer (host) stands in for the listener — a newcomer, a curious colleague, a smart generalist. They ask the questions your reader would ask, define the acronyms, summarize the turns.
- The expert (guest) is you — the author of the original article, delivering the proprietary insights, the data, the opinion.
This isn't a stylistic choice. It mirrors the Socratic method, and it tricks the brain into treating the episode as a private mentoring session instead of a corporate lecture. Trust forms faster, retention goes up, and shareability quietly improves because conversations are easier to recommend than essays.
The 15-minute workflow
Here's the exact path from a published Medium article to a Spotify-ready episode, using Sprep as the audio platform.
1. Set the strategic guardrails (3 min)
Open Sprep and go to Style Configuration before you upload anything. The AI needs to know what it's writing for, not just what it's writing about.
- Format: "Interview (Q&A)" — this locks in the two-speaker dynamic.
- Intent and audience: "Educate / Explain" for a "Beginner / New Hire" audience makes the host naturally pause to ask clarifying questions when a section gets dense.
- Tone: "Professional / Corporate" for B2B, "Conversational / Casual" for a personal brand.
This step is the one most people skip, and it's the reason most AI podcasts sound generic. The guardrails are the show.
2. Curate the narrative flow (2 min)
Paste the Medium text (or upload it as a PDF). Sprep parses the document and surfaces a visual table of contents — each section becomes a drag-and-drop block.
This is where you stop being a writer and start being a producer. A Medium article is structured for scanning; a podcast needs a hook. Drag the conclusion or the strongest statistic to the top to open with tension, then let the rest of the structure pay it off. Reordering takes seconds and changes the energy of the whole episode.
3. Edit the dialogue (3 min)
Sprep generates the full transcript as a chat interface — host on one side, expert on the other. This is the Human-in-the-Loop step that separates a publishable asset from a generic AI clip.
- Rewrite lines that miss your voice.
- Spell tricky acronyms phonetically so the voice model nails them.
- Add a spoken call-to-action at the end: "For the charts we discussed, the full Medium article is in the show notes."
Every word that ships gets reviewed by a human before any audio is rendered. That's the only reason you'd put this episode under your name.
4. Direct the voices and the emotion (4 min)
Open the voice panel and assign personalities. Sprep tags voices by trait — pick a youthful, curious voice for the host and a mature, authoritative one for the expert. The contrast is what sells the dialogue.
Then add prosody cues in brackets next to specific lines: [serious], [excited], [curious]. The model adapts pitch, pace and breath for that line only. It's the same thing a podcast director would whisper into a guest's headphones — and it's the difference between "this sounds like AI" and "wait, is this AI?"
5. Translate, if you have international readers (optional)
Pick a target language from the dropdown — Hindi, German, French, Spanish — and Sprep replaces the script with a localized version while preserving the pacing and emotional cues. You approve once in the source language; you ship in five.
6. Generate, export, distribute (3 min)
Hit generate, preview the track, download the MP3. Upload to Spotify for Podcasters (or Transistor, Buzzsprout, your existing host). Paste your Medium intro paragraph into the episode description for search, and publish — your RSS fans out to Spotify, Apple Podcasts and the rest automatically.
Total elapsed: about 15 minutes.
What this unlocks
Embedding a 10-minute Spotify player at the top of your Medium article gives your readers a choice they don't currently have. The ones who would have finished the text still finish it. The ones who wouldn't — the commuters, the parents, the people in back-to-back meetings — now have a path.
Two outcomes show up almost immediately:
- Retention goes up. Text retention sits around 10–20% because of skimming. Narrative audio routinely lands closer to 80–95% because listeners default to consuming a track end-to-end.
- You capture dark social. Podcasts get shared in DMs, Slack threads and WhatsApp chats far more readily than gated PDFs. You don't pay more to acquire those listeners — they were already in your funnel, just bouncing.
Stop letting your best thinking die in the "read later" graveyard. Repurpose it as a conversation, ship it to where your audience already listens, and let the format do the distribution work.
This walkthrough is adapted with thanks from Ben's original Medium piece, How to Turn Your Medium Article into a Spotify Podcast in 15 Minutes.
FAQ
Frequently asked questions
Why turn a Medium article into a podcast at all?
Why a two-host format instead of a single narrator?
How long does this actually take end-to-end?
Can I publish in more than one language?
Where does the final episode get hosted?
See it in action
Convert your own documents into podcasts