By whispertranscribe.ai Editorial Team · Last updated: November 2026 · 12 min read
The right podcast transcript format depends on where the text will live. On your episode page, use readable paragraphs with a speaker name before every turn and occasional timestamps. For Apple Podcasts, Spotify, and your RSS feed, use a timed WebVTT (.vtt) file, with SRT as the fallback. Keep a plain TXT or DOCX copy for editing.
Below: four copy-ready templates, the file rules from the Podcasting 2.0 spec, Apple, Spotify, and YouTube (checked November 2026), and a publishing checklist.
Quick answer: which podcast transcript format should you use?
Most shows need two versions of the same words: a reading version for people and a timed file for apps. Pick the reading version by audience and the timed file by platform.
| Format | Best for | Pros | Cons | Where it's accepted |
|---|---|---|---|---|
| Plain text (TXT) | Editing, show notes, repurposing into posts | Opens anywhere; easy to search and paste | No timing; speaker names only if you type them | Your site; RSS tag as text/plain |
| Speaker-labeled document (DOCX or web page) | Episode pages, guest review, accessibility | Easiest to read; supports headings and links | No sync with the audio | Your site; RSS tag as text/html |
| Timestamped text | Long interviews, quote checks, research | Readers can jump to a moment | Too many stamps clutter the page | Your site |
| SRT (.srt) | Upload to Spotify or YouTube; caption-style feeds | Simple and widely supported | No voice tag, so names go inside the text | Apple (subscriber upload), Spotify, YouTube, RSS tag |
| WebVTT (.vtt) | Apple Podcasts, Spotify, web players | Speaker names via voice tags; browsers read it natively | Strict syntax; a stray line breaks it | Apple (feed and subscriber upload), Spotify, YouTube, RSS tag |
| JSON (Podcasting 2.0) | Apps that highlight word by word | Most precise timing | Few hosts generate it | RSS tag as application/json |
Rule of thumb: if you can only make one timed file, make it VTT. The Podcasting 2.0 transcript documentation recommends WebVTT for exactly that case, noting that Apple Podcasts uses it for ingest and that browsers support it natively.
What a good podcast transcript includes
A good podcast transcript lets someone who never presses play follow the whole episode. The W3C describes a podcast transcript as simple text paragraphs with the speakers identified, plus the non-speech audio a listener would need to understand the content.
In practice that comes down to five habits:
- Name the speaker at every turn. Use the full name once near the top, then a short form. Never mix "Mia", "Host", and "Speaker 1" in one file.
- Start a new paragraph at every change of speaker. Split long answers where the topic shifts.
- Put sounds that carry meaning in square brackets.
[laughs],[phone rings],[theme music]. Skip background noise nobody needs. - Mark anything you added. The W3C notes that added clarifications are fine as long as readers can tell they aren't part of the audio, for example by using brackets.
- Use timestamps sparingly. W3C guidance says to include them only when useful, and they rarely need end times.
Clean or verbatim? For a public episode page, clean verbatim works best: keep every meaningful word and drop "um", stutters, and false starts. Keep full verbatim when exact wording matters, such as research interviews or quotes that might be disputed. Whichever you choose, say so in one line at the top of the transcript.
Copy-ready podcast transcript templates
Each template below is a fictional example we wrote for this guide (the show "Trailhead Radio" doesn't exist). Replace the bracketed placeholders and keep the structure.
Template 1: Interview format (two or more speakers)
Use this for guest episodes and panels. Bold the labels if your site supports it.
EXAMPLE (fictional show)
Trailhead Radio, Episode 58: Packing for a First Overnight Hike
Host: Mia Chen. Guest: Luis Ortega, park ranger.
Transcript style: clean verbatim.
[theme music]
Mia Chen: Welcome back to Trailhead Radio. I'm Mia, and today I'm
talking to ranger Luis Ortega about the mistakes he sees on first
overnight trips.
Luis Ortega: Thanks for having me. The big one is weight. People pack
for every possible problem and end up too tired to enjoy the trail.
Mia: So where do you start?
Luis: With water. Check where you can refill before you leave, then
carry only what gets you to the next source. [laughs] My first trip,
I carried six liters up a creek trail.
Template 2: Solo or monologue episode
Solo episodes have no speaker changes, so structure comes from headings and paragraphs instead of labels.
EXAMPLE (fictional show)
Trailhead Radio, Episode 59: Reading a Trail Map
Speaker: Mia Chen. Transcript style: clean verbatim.
Introduction
Today is a short one. I want to walk you through the three things I
check on a trail map before every hike, in the order I check them.
1. Contour lines
The closer the lines, the steeper the ground.
2. Water sources
Note the distance between the blue symbols along the route.
[Sponsor message]
3. Exit points
Find at least one way off the trail before the halfway mark.
Template 3: Timestamped transcript
Use this for long interviews where readers need to find a moment fast. A stamp at each turn, or every few minutes, is plenty.
EXAMPLE (fictional show)
Trailhead Radio, Episode 58 (running time 41:20)
[00:00] [theme music]
[00:12] Mia Chen: Welcome back to Trailhead Radio. Today I'm talking
to ranger Luis Ortega about first overnight trips.
[00:21] Luis Ortega: Thanks for having me. The big mistake is weight.
[03:47] Mia: Let's talk about food. How much do people overpack?
[03:52] Luis: Almost always by a full day. [laughs]
[17:05] [inaudible 17:05] Luis: ...and that's why I carry a paper map.
For episodes over an hour, write the hours too ([01:03:47]) so every stamp has the same shape.
Template 4: Chaptered transcript for long episodes
Chapters help readers and screen reader users jump through a long conversation. W3C guidance suggests adding headings where they make a transcript more usable.
<!-- EXAMPLE (fictional show) -->
## Transcript: Packing for a First Overnight Hike
**Episode 58 · Host: Mia Chen · Guest: Luis Ortega · Clean verbatim**
### Chapter 1 (00:00): Why first trips go wrong
**Mia Chen:** Welcome back to Trailhead Radio...
**Luis Ortega:** The big one is weight...
### Chapter 2 (03:47): Food and water
**Mia:** Let's talk about food...
### Chapter 3 (17:05): Navigation without a signal
**Luis:** I always carry a paper map...
### Resources mentioned
- [Link to the park's trail conditions page]
Use the same chapter times as your show notes, so both lists agree.
Podcast transcript file formats: VTT, SRT, JSON, and HTML
Timed files follow stricter rules than reading copies. Here's the same fictional exchange in each.
WebVTT (.vtt): the safest timed file
A VTT file starts with WEBVTT, uses a period before the milliseconds, and can name the speaker in a voice tag. Apple says that providing a VTT file lets you identify the speaker on each line, and the name then appears in its transcript view whenever the speaker changes.
WEBVTT
00:00:12.000 --> 00:00:17.400
<v Mia Chen>Welcome back to Trailhead Radio.
00:00:17.400 --> 00:00:21.000
<v Mia Chen>Today I'm talking to ranger Luis Ortega.
00:00:21.000 --> 00:00:24.800
<v Luis Ortega>Thanks for having me.
The big mistake is weight.
The Podcasting 2.0 docs suggest keeping VTT lines to 65 characters or fewer so the file also works as captions.
SRT (.srt): the fallback
SRT numbers each cue, uses a comma before the milliseconds, and has no voice tag. The Podcasting 2.0 docs set a podcast SRT profile: at most two lines per cue, at most 32 characters per line, and a new cue whenever the speaker changes, starting with the name and a colon.
1
00:00:12,000 --> 00:00:15,000
Mia: Welcome back
to Trailhead Radio.
2
00:00:21,000 --> 00:00:24,800
Luis: Thanks for having me.
The big mistake is weight.
New to SRT? Our explainer on what an SRT file is walks through every line.
JSON and HTML (Podcasting 2.0)
The spec also defines two less common formats. JSON stores a spec version plus an array of segments; every segment carries speaker, startTime, endTime, and body fields, and one segment per word allows word-by-word highlighting. HTML suits transcripts with little or no timing: the speaker's name sits in <cite>, the start in <time>, and the words in <p>. The spec calls it low-fidelity, meant as an accessibility aid and searchable text.
How to add a transcript to your RSS feed (Podcasting 2.0 tag)
Podcast apps find your own transcript through the <podcast:transcript> tag inside each episode's <item>. Most hosts write this tag for you when you attach a file, so check your host's dashboard before you edit a feed by hand.
According to the official tag documentation:
url(required): where the transcript file lives.type(required): the MIME type. The docs listtext/plain,text/html,text/vtt,application/json, andapplication/x-subrip(SRT).language(optional): if left out, apps assume the language in your feed's<language>element.rel="captions"(optional): tells apps the file is closed captions with time codes, whatever its MIME type.- Several tags are allowed, so you can offer one file per format.
<!-- EXAMPLE: one episode offering three transcript files -->
<rss version="2.0" xmlns:podcast="https://podcastindex.org/namespace/1.0">
<channel>
<language>en</language>
<item>
<title>Packing for a First Overnight Hike</title>
<podcast:transcript url="https://example.com/ep58/transcript.vtt"
type="text/vtt" />
<podcast:transcript url="https://example.com/ep58/transcript.srt"
type="application/x-subrip" rel="captions" />
<podcast:transcript url="https://example.com/ep58/transcript.html"
type="text/html" />
</item>
</channel>
</rss>
One easily missed detail: the transcript format docs say browser-based players need CORS headers on the file's server to load it from another website. If a web player shows no transcript while Apple does, check that first.
Apple Podcasts, Spotify, and YouTube transcript requirements
Each platform takes timed files, but each gets them a different way. Details below were checked on each platform's help page in November 2026.
| Platform | Files accepted | How your file gets there | Watch out for | Official source |
|---|---|---|---|---|
| Apple Podcasts | VTT or SRT | RSS transcript tag; direct upload in Apple Podcasts Connect for subscriber episodes | You must choose "Display transcripts I provide"; 11 supported languages | Apple Podcasts for Creators |
| Spotify | VTT or SRT, up to 5MB | Upload in Spotify for Creators (eligible shows) | Turning on auto-generated transcripts replaces your upload | Spotify for Creators Help |
| YouTube | SRT, VTT, and other caption formats | YouTube Studio, Subtitles, Upload file | SRT must be plain UTF-8 with no markup | YouTube Help |
| Other apps | Depends on the app | RSS transcript tag | Not every app reads the tag | Podcasting 2.0 docs |
Apple Podcasts
- Apple creates transcripts automatically for new episodes. To show yours, change the transcript setting on the show's Availability tab in Apple Podcasts Connect, or use a custom setting for one episode.
- Transcripts work on iOS 17.4 or later in English, Danish, Dutch, Finnish, French, German, Italian, Norwegian, Portuguese, Spanish, and Swedish. For other languages, Apple suggests linking the transcript in your show notes.
- Your file must be VTT or SRT, free of spelling and punctuation errors, and must cover all segments of the episode. Files that fail Apple's quality standards aren't displayed.
- Apple doesn't show segments changed by dynamically inserted audio, doesn't show music lyrics, and may not transcribe episodes over 10 hours.
- The app labels your file "Provided by" and its own "Automatically generated."
Spotify
- Transcripts aren't available to every creator. If your show is eligible, go to Episodes, select the episode, click Details, then Edit under Transcripts. From there you can download Spotify's VTT, fix it, and upload your own file.
- Spotify asks for VTT or SRT, a maximum of 5MB, and timestamps, without which the transcript won't sync during playback.
- A setting called Distribute transcripts across other platforms via RSS puts the transcripts in your feed for other apps.
YouTube
Video podcasts on YouTube use ordinary caption tracks. In YouTube Studio, choose Subtitles, select the video, add a language, then Upload file and pick With timing or Without timing. YouTube's help page advises against relying on untimed transcripts when a video runs past an hour or the audio is poor, which describes most full-length episodes, so upload a timed SRT or VTT file instead.
Accessibility: what WCAG expects from a podcast transcript
This section is general information, not legal advice. Whether a law applies to your show depends on your country and organization.
The Web Content Accessibility Guidelines treat a podcast as audio-only media. Under Success Criterion 1.2.1, prerecorded audio-only content needs a text alternative that "presents equivalent information." The W3C's transcripts guide adds that transcripts for prerecorded audio are required at Level A, while transcripts for live audio-only content are at Level AAA.
What that means for format:
- Equivalent means complete. Include all speech plus meaningful sounds, not a summary.
- It should be real text, so screen readers, braille displays, and search tools can use it.
- Make it easy to find. The W3C recommends putting the transcript, or a link to it, right next to the media, and usually on the same page.
- A caption file isn't a reading transcript. If you start from SRT or VTT, merge the short cues into paragraphs, which is what the W3C's own example does.
Do podcast transcripts help SEO?
They can, but be careful with promises. Here's what the sources actually say:
- The W3C notes that adding headings and links to a transcript makes it more usable and "also helps with SEO."
- Spotify says its transcripts help it understand what a show is about, and they power discovery features such as chapters and guest and topic tags (Spotify for Creators).
- Apple lets listeners search an episode for a word or phrase and tap the text to play from that point.
None of these promises higher rankings. Publish the transcript because it makes the episode usable for more people, and treat any traffic gain as a bonus.
How to create a podcast transcript with whispertranscribe.ai
whispertranscribe.ai is a browser-based transcription tool powered by OpenAI's Whisper model. It turns an episode into text you can export in the formats above. It works in a web browser; there's no app to install.
- Open the podcast transcript generator and choose the spoken language from 130+ languages. For rough recordings, many languages have a Noise / Low-Quality Audio/Video option, and there's a Multilingual Mix option for episodes that switch languages.
- Add the episode. Upload the audio or video file (MP3, WAV, M4A, MP4, and more), up to 5GB and 10 hours per file. Or use Transcribe from URL, which accepts YouTube, Vimeo, Google Drive, and Dropbox links. That page doesn't list Apple Podcasts or Spotify links or RSS feeds, so for an audio-only episode, upload the file from your host or editing software.
- Turn on Speaker Detection for interviews and panels.
- Review the draft. Accuracy can reach up to 99% on clear audio, but crosstalk, music under voices, noise, and strong accents lower it. Check names, numbers, and the first line after each speaker change.
- Export what each destination needs: TXT or DOCX for your episode page and show notes, VTT or SRT for Apple Podcasts, Spotify, and YouTube, plus PDF or CSV if you need them.
If the export uses generic labels, replace them with real names using find and replace before you publish. To make subtitles for clips, the AI subtitle generator has a maximum-words-per-subtitle setting, and MP3 to SRT handles audio-only files. For interview-heavy shows, see free interview transcription. Need a smaller file first? Use convert audio or video to MP3.
You can start without creating an account; check that page for what free use currently includes. Paid plans start from $9.99/month billed annually (pricing). All tools are listed on the features page.
Podcast transcript publishing checklist
- One house style for labels, brackets, and timestamps, stated at the top
- Every speaker named, with spellings matching the episode description
- Names, numbers, and brand terms checked against the audio
- Reading version published next to the player, or linked from it
- Timed file saved as UTF-8,
.vttor.srt, not.vtt.txt - VTT starts with
WEBVTT; SRT timestamps use commas - File covers the whole episode, including intro and outro
- RSS tag has the right
type, and the file URL opens publicly - Apple set to "Display transcripts I provide"
- Spotify upload is under 5MB and auto-generated transcripts stay off
- Checked the live episode in each app 24 hours after publishing
Troubleshooting: common podcast transcript format problems
| What you see | Likely cause | Fix |
|---|---|---|
| Apple still labels the transcript "Automatically generated" | Transcript setting unchanged | Select "Display transcripts I provide" for the show or the episode |
| No transcript at all on Apple | Unsupported language, explicit episode with restrictions on, or under 24 hours since publishing | Check the language and explicit settings, then wait a day |
| Parts of the episode missing on Apple | Dynamically inserted audio changed after transcription | Expected behavior; keep the full text on your site |
| No speaker names on Apple | File is SRT or has no voice tags | Export VTT and add <v Name> tags |
| Spotify rejects the upload | Wrong file type or over 5MB | Export VTT or SRT; trim long cue text or split notes out |
| Spotify transcript doesn't follow the audio | No timestamps in the file | Upload a timed VTT or SRT, not TXT |
| Your upload vanished on Spotify | Auto-generated transcripts were turned on | Turn them off and upload again |
| Web player shows nothing, Apple works | Missing CORS headers on the file host | Ask your host or CDN to allow cross-origin requests |
| Accented letters look broken | File not saved as UTF-8 | Re-save as UTF-8 in a plain text editor |
| Readers find the page hard to follow | Caption file pasted as the transcript | Merge cues into paragraphs and add speaker labels |
More caption-file fixes are in our guide to exporting SRT subtitles.
FAQs
What is the best format for a podcast transcript?
There are really two best formats, one for people and one for apps. For readers, use clean verbatim text with the speaker's name before each turn, a new paragraph at every speaker change, and sounds in square brackets. For podcast apps, use a timed WebVTT file, which Apple Podcasts and Spotify accept and the Podcasting 2.0 docs recommend if you can only support one format. SRT works as a backup.
Should a podcast transcript include timestamps?
Your timed file must have them, but your reading version only needs a few. Spotify says an uploaded transcript without timestamps won't sync during playback, and Apple expects VTT or SRT, both of which are timed. On a web page, W3C guidance says to include timestamps only when they're useful. A stamp at each speaker turn, or at each chapter, lets readers find a moment without cluttering the text.
What file format does Spotify accept for podcast transcripts?
Spotify accepts VTT or SRT files up to 5MB, uploaded in Spotify for Creators under the episode's Details and Transcripts section. The file needs timestamps, or it won't sync during playback. Transcripts aren't available to every creator, so the option may not appear for your show. If you later turn on Spotify's auto-generated transcripts, they replace the file you uploaded.
How do I add a transcript to my podcast RSS feed?
Add a podcast:transcript tag inside the episode's item element, with the file's URL and its MIME type, such as text/vtt for WebVTT or application/x-subrip for SRT. You can add several tags for different formats, add a language attribute, and use rel="captions" for timed caption files. Most podcast hosts write this tag for you when you attach a transcript, so check your host's dashboard first.
Do podcasts need transcripts for accessibility?
Under WCAG, prerecorded audio-only content such as a podcast episode needs a text alternative that presents equivalent information, and the W3C lists this at Level A, the most basic conformance level. Whether WCAG is legally required for your show depends on your country, sector, and organization. This is general information, not legal advice, so check the rules that apply to you.
Can I turn an SRT or VTT file into a readable transcript?
Yes. Remove the cue numbers and timing lines, merge the short caption lines into paragraphs, and start a new paragraph whenever the speaker changes. Turn VTT voice tags into names followed by a colon, and keep bracketed sounds that matter. The W3C uses the same approach in its own example, grouping several caption lines into each paragraph of the transcript.
Ready to format your next episode? Upload it to the whispertranscribe.ai podcast transcript generator and export TXT, DOCX, VTT, or SRT in 130+ languages, with speaker detection.
Related guides: What is an SRT file? · How to export SRT subtitles · How to transcribe MP3 to text · Subtitles vs captions
References
- Podcast Index: podcast:transcript tag and Transcript File Format Details
- Apple Podcasts for Creators: Transcripts on Apple Podcasts
- Spotify for Creators: Managing episode transcripts on Spotify and automatic transcripts, chapters, and more
- YouTube Help: Supported subtitle and closed caption files and Add subtitles & captions
- W3C WAI: Transcripts
- W3C: Understanding SC 1.2.1 Audio-only and Video-only (Prerecorded)
