How to make a podcast trailer: length, script, AI voice
To make a podcast trailer, script an intro, highlights, a hook and a call to follow, keep it to two minutes or less, and mix the voice over a music bed.
To make a podcast trailer, write a short script that introduces the show and its hosts, shares a few highlights, leaves a question open, and ends with a call to follow; Apple's podcaster guide says to keep trailers to two minutes or less. Then record or voice the script, mix a music bed under it, and export one audio file for your feed, plus a square or vertical video version for social.
The trailer advice comes from Apple Podcasts for Creators, Make a great podcast trailer, read on 2026-09-29. The build steps follow Sume's Music Router, Timeline 1.0 and Audio detach docs and the TTS schema in the Sume API reference; anything called current behavior is read from Sume's code.
What is a podcast trailer?
A short audio or video piece that sells the show before someone commits to an episode. Apple's guide names three kinds:
- Coming soon trailer: teases a brand-new show before launch, or a new season or series.
- Show trailer: the podcast's highlight reel, used wherever you introduce the show, such as social channels and your website.
- Teaser trailer: played at the end of an episode to preview the next one.
How long should a podcast trailer be, and what goes in the script?
Apple says to keep trailers to two minutes or less, and to start from an outline so you stay inside that limit. Its standard show-trailer outline has four parts. The sample lines are ours, for a made-up show; replace them with your own.
Apple also suggests including the sounds listeners connect with your show, such as recurring background music or a catchphrase.
| Part | What Apple says to include | Sample line |
|---|---|---|
| 1. Introduce your show | The show and hosts, then a brief explanation of what it is about | “This is Night Shift, a show about people who work while the city sleeps. I'm your host, Sam.” |
| 2. Share the highlights | The moments that define the listening experience; favorite guest quotes | Two or three short quotes, or a line describing each story. |
| 3. Make them want more | Tension or curiosity: a problem, a twist, or an unanswered question | “And then the power went out on the whole block.” |
| 4. Give them a call to action | When new episodes arrive, and a reminder to follow the show | “New episodes every Tuesday. Follow Night Shift so you don't miss one.” |
How do I make a podcast trailer with AI?
A coming-soon trailer, where no episode exists yet, is a narrated script over music, which you can build end to end with a few API calls. How to mix voice with background music shows the mix and export requests in full.
- Voice: send the script to
POST /v1/tts-1.0/generate. One request takes up to 20,000 characters; setlanguagefor a non-English script, since omitted it defaults to English. - Music: send a brief to
POST /v1/music-router/generate. The request has no duration field, so put the length in the prompt, and the music docs suggest closing with “Instrumental, no vocals.” plus “no spoken word” under narration. - Mix and export: one Timeline 1.0 render ducks the music under the voice (it needs at least one clip in
video[], such as the cover-art clip below), thenPOST /v1/audio-detachturns the render into a WAV (the default) or a 128 kbps MP3 for your podcast host.
Can I add highlights from my recorded episodes?
Plan to cut those in your own editor. Timeline, audio detach and the other media tools read only your workspace's media.sume.com files, such as the outputs of earlier Sume jobs, so episodes recorded elsewhere can't go into these steps. Episodes you generated on Sume, for example with an AI podcast generator from text, are already Sume files and can be sliced and reused.
How do I make a video version for social?
Apple notes that a square-cropped video may work best for an Instagram feed, while Instagram Stories need a vertical video. In the same Timeline render, set output.width and output.height to even numbers from 256 to 2160, such as 1080×1080 or 1080×1920. For the picture, put your cover art at a public HTTPS URL and animate it as the first_frame of a POST /v1/videos clip; in current code the render drops each clip's own audio, so the trailer mix stays clean. A slot longer than its clip is padded or looped, which the render reports as a warning, so use several clips for a long trailer. Podcast cover art size covers the artwork itself.
How much does an AI podcast trailer cost?
Each step is billed per call, plus a 5.5% agent fee by default:
- Voice: $0.0475 per 1,000 characters, spaces and punctuation included.
- Music bed: $0.125 per audio.
- Mix: $0.10 per output minute, so a trailer of two minutes or less bills at most two minutes.
- Audio file: $0.01 per job.
- Cover-art clips: by model, at provider list × 1.25; each model's rates are under
pricing_skusonGET /v1/videos/models.
Sources
Related posts
More in Use cases
- How to make AI cat videos: one cat, many shots
To make AI cat videos, approve one still of your cat, start every shot from it, and join the shots under narration. Dancing and talking cats, costs.
- How to make an AI movie: shot by shot, then the cut
You make an AI movie shot by shot: clips of up to 30 seconds, generated from a shot list, checked, voiced, then cut together. Workflow, limits, costs.
- Law firm video marketing with an AI avatar of the attorney
Law firm video marketing with AI: short attorney intro and practice-area clips from an avatar of the lawyer, what not to generate, how it's made and billed.
- LinkedIn content credentials: what the CR label means
LinkedIn shows a C2PA icon on images and videos signed with Content Credentials; clicking it shows whether AI was used, the tool, and who signed it.
Written by Sume