Podcast Clip Maker: How to Turn One Episode Into YouTube Shorts

A podcast clip maker takes one long-form episode — a 90-minute conversation, a YouTube interview — finds the strong moments, and cuts each into a 9:16 vertical Short with word-by-word captions burned in. You upload the episode, review the clips it stages for you, and publish the good ones to YouTube Shorts. One episode becomes a handful of vertical clips.

That is the whole job. The rest of this article shows you the input, the output, and every step in between — so you can picture the finished clip before you sign up for anything. Everpop is the tool being shown, but the workflow is the same idea any serious creator should expect from AI content repurposing: honest input, a real clip you can see, and a way to check what happened after you post.

What does a podcast clip maker actually do?

You give it a long recording. It gives you back short vertical clips ready for Shorts. (Most podcasts still post Shorts only sporadically — we measured 940 channels.) In between, three things happen that are worth understanding, because they decide whether the output is usable or landfill.

First, it finds moments. Instead of you scrubbing a two-hour timeline hunting for the one story that lands, the engine reads the whole episode and picks the segments with a strong opening and a payoff. Second, it reframes. A podcast is usually shot wide or 16:9; a Short is tall. The clip is rendered at 9:16 so it fills a phone screen with no black bars. Third, it captions. Most podcast audio is people talking, and on a muted mobile feed nobody reads a clip without captions — so the words are burned directly into the video, synced to the audio.

With Everpop, the input constraint is the honest part: the source video comes from you. You drag-drop the episode into the upload hub, or drop the finished file into a shared Google Drive folder. Pasting a YouTube link is refused — the tool never downloads from YouTube. Your file comes from you.

What will the finished Short actually look like?

This is the part people skip and shouldn't. Here is the concrete artifact.

The clip is a vertical 9:16 MP4, H.264 video and AAC audio, rendered at 1080×1920 — HD, Everpop's render spec for 9:16 Shorts. YouTube caps Short uploads at 1080p and lets vertical videos up to three minutes qualify as Shorts, so a 1080×1920 clip lands exactly in-spec (YouTube Help).

On top of the video sit word-by-word animated captions, burned in — not soft subtitles a viewer can toggle off, but part of the picture, each word popping in as it's spoken. You pick from three caption presets:

• Pop (the default) — white text that sweeps to gold as each word is hit, with a pop-in.
• Boxed — words on a solid block for maximum contrast.
• Minimal — clean type, no ornament.

Captions can be set in six curated fonts, free on every plan: Anton (the default), Archivo Black, Bebas Neue, Bangers, Titan One, and Passion One. The captions are positioned to sit above the Shorts interface — the like, comment and title chrome that crowds the bottom and right of the screen — so your words don't get buried under buttons.

The clip opens with a hook and can be built as a montage of several cuts rather than one flat take. And you choose the length up front with a set of pills: Auto, 15, 30, 45, 60, or 90 seconds (roughly 20 seconds minimum). Auto lets the engine size the clip to the moment; the fixed pills let you standardize.

So the before-and-after is concrete: one 90-minute episode in, a handful of captioned vertical clips out — each one a finished Short you could post as-is.

How do I go from raw episode to posted Short?

Here is the actual flow, start to finish. (For the same flow across the whole product, see the step-by-step walkthrough.)

Step	What you do	What Everpop does
1. Add the video	Drag-drop the episode into the upload hub, or drop it in your Drive Finals folder	Ingests the file (resumable for large uploads), then clips it automatically
2. Wait	Nothing	Finds strong moments, renders 9:16 clips, burns captions, runs quality-control checks
3. Review	Open the Clips page, watch each card, thumbs up/down	Stages every clip for your approval — nothing posts until you say so
4. Re-render (optional)	Not happy with a clip? Hit Re-render	Rebuilds it — the first 3 re-renders per clip are free
5. Publish	Click Publish, set privacy	Posts natively to your connected YouTube channel as a Short
6. Read the receipt	Come back at 48h and 7d	Pulls the real numbers from YouTube and shows you a receipt

The review step is the default, not an afterthought. On the Clips page you see a grid of 9:16 cards. Each card shows the aspect ratio, a thumbs up/down rating, and buttons to Publish, Re-render, or Discard. Filter chips across the top — All, Ready, Published, Processing, Failed — let you find clips fast, and you can search by title. Nothing publishes behind your back. (If you'd rather it did, an optional Autopilot mode per channel skips review and posts rendered clips straight to your destinations.)

This is what separates a real AI Shorts generator from a render button that dumps files on you: you get to see and reject before anything hits your channel.

What does the receipt tell me after I post?

After a clip goes live, Everpop pulls the real performance data from YouTube — the official Analytics API, authenticated as you — at two checkpoints: 48 hours and 7 days after publish.

A receipt shows views, average view %, subscribers gained, likes, and watch minutes for that specific clip. While it's still measuring, the card shows a "measuring in Xh" state instead of pretending it has numbers it doesn't. Each receipt is signed and shareable through a public link, so if you're pitching a sponsor or an agency, you can hand them proof of what a clip did — not a screenshot anyone could fake.

That's the honest boundary worth stating plainly: a receipt proves what happened. It is not a prediction, and no clip tool can promise you a hit. What the receipt does do — beyond showing you the truth — is feed a per-channel learning loop, so which moments get picked keeps tuning to what actually performs on your channel.

Which platforms can I publish to?

Today, YouTube Shorts is native: clips publish directly to your connected channel through the official API, and you set the privacy at publish time. There's also a scheduler — a "Go live" picker with a horizon from two minutes to seven days out — so you can queue clips instead of posting them all at once.

TikTok is a first-party integration currently in pilot. Instagram Reels and Facebook Reels are coming soon. This piece of AI video clipping software is honest about the map: YouTube is live, the rest is on the way.

If you'd rather take a clip into Premiere or DaVinci Resolve, each clip can export as a .zip with an FCPXML timeline, an EDL, an SRT caption file, and a README — so the automated cut becomes a starting point, not a dead end.