SRT Files for Podcast Clips: A Practical Guide

An SRT file is a plain-text file that pairs lines of caption text with the exact moments they should appear on screen. Each block, called a cue, has four parts: a number, a start-and-end timestamp, the caption text, and a blank line. That's the whole format. You can open one in Notepad, read it, and fix it by hand, which is the single most useful thing to know when a subtitle file breaks the night before you post.

For podcast clips, SRT matters because it's the universal handoff format. Your transcription tool exports one, your editor imports one, YouTube ingests one, and almost every captioning app speaks it. Understanding the four-part structure means you stop treating the file as a black box and start treating it as text you can edit. Below: a real cue read line by line, the SRT-vs-VTT call for clips, and the four ways these files break, each with the two-minute fix.

What is an SRT file?

An SRT file (SubRip Subtitle, extension .srt) is a text file that holds timed captions. It lists numbered cues, each saying show this text from this second to that second. It carries no styling, no font, color, or position, only words and timing. Any video tool can read it, which is why it's the default way subtitles move between apps.

The name comes from SubRip, an early-2000s program that ripped subtitles off DVDs. The tool faded; the format won. Today SRT is the closest thing captioning has to a universal language, the way .mp3 is for audio. When a podcast host exports a transcript "as subtitles," you almost always get an SRT.

What it is not: a styled caption. The bold, animated, brand-colored captions you see on a Reel are rendered (burned in) from an SRT or its data, but the SRT itself is bare text plus timecodes. The look, your caption font, color, and position, gets applied later, in your editor or on the platform, not in the file. That separation is the whole point, and the reason the format is so portable: the same SRT can drive a dozen different visual styles.

The anatomy of one SRT cue (worked example)

Here is a single, real cue. Read it top to bottom and the format stops being mysterious:

``srt 12 00:00:24,500 --> 00:00:27,000 And that's the part nobody tells you about quitting your job. ``

Four parts, in this exact order:

1. The sequence number (12). Cues count up from 1, in order. It's mostly bookkeeping, most players ignore the actual value and just read top to bottom, but skipped or out-of-order numbers are a classic sign a file got edited carelessly.
2. The timestamp range (00:00:24,500 --> 00:00:27,000). Start time, a space, an arrow (-->), a space, end time. Each timestamp is hours:minutes:seconds,milliseconds. The detail that trips everyone: the milliseconds are separated by a comma, not a period. This caption appears at 24.5 seconds and leaves at 27 seconds.
3. The caption text (one or two lines). Keep it to one or two short lines per cue, that's all that fits comfortably on a phone. Plain text only.
4. The blank line. One empty line ends the cue and tells the parser the next number begins a new one. Forget it and the file collapses into one giant unreadable cue.

That's the entire specification you need. If you can read those four parts, you can read, sanity-check, and repair any SRT a tool hands you.

How do you create an SRT file?

You create an SRT one of three ways: export one from a transcription or captioning tool (fastest), generate it with your video editor's auto-subtitle feature, or type it by hand in a plain-text editor for a short clip. For anything past a single cue, never type the timecodes by hand, getting the comma, the arrow spacing, and the millisecond format right manually is exactly where files break.

The practical route for a show is to transcribe the full episode once, export an SRT, and reuse it. When you cut a clip, you only need the cues that fall inside that clip's time window, with the timestamps shifted so the clip starts at zero. Good clipping tools do that shift automatically. If you're doing it by hand, that re-timing step is the tedious part, and the reason most people let software handle it.

If you want the bigger picture of how transcription, styling, and burn-in fit together, the end-to-end workflow for adding captions to podcast clips covers all five steps; this guide zooms in on the file format underneath them. And if you're weighing whether to trust the auto output or fix it yourself, auto vs manual captions: which is worth it makes the call.

SRT vs VTT: which should you use?

Use SRT when you need maximum compatibility, uploading to YouTube, importing into a desktop editor, or handing a file to a clipping tool. Use WebVTT (.vtt) when captions play in a web browser via HTML5 video or when you need styling and positioning baked into the file. For most podcast clips posted natively to social, you burn the captions in and the file format barely matters; the SRT is just the intermediate.

The two formats are siblings. VTT was built as the web-native successor to SRT and the structures look nearly identical, with three differences that actually matter:

• The millisecond separator. SRT uses a comma (00:00:27,000); VTT uses a period (00:00:27.000). This one character is why a renamed file fails to load, see the fixes below.
• The header. A VTT file must start with the line WEBVTT. An SRT has no header; it begins straight at cue number 1.
• Styling. VTT can carry positioning, alignment, and basic styling cues inline. SRT cannot, it's pure text, which is exactly why it travels so well.

The decision rule for clips is short. Posting to Shorts, Reels, or TikTok? You'll burn the captions in, see burned-in vs soft captions for why hardcoded is the standard for native social, so export whatever your tool gives you and move on. Uploading a long video to YouTube and want a separate caption track? SRT, every time, YouTube reads it cleanly and you keep the source file. Embedding video on your own site? VTT, because the browser's native player expects it.

Why the file format is worth knowing at all

Captions are not optional on social, which is why the format under them is worth ten minutes of your attention. A widely cited estimate puts around 85% of social video watched with the sound off (Digiday, 2016, attributed to multiple publishers), treat that as directional, not precise: it's a decade-old, self-reported figure, and viewing habits have shifted since. The exact share drifts; the behavior doesn't. Most people meet your show on mute, reading.

That hits clips harder than any other format, because clips are often the front door to the whole show. One studio's client data attributes 20–40% of new audience and reach lifts of 2–5× to clips for video shows (Podcast Studio Glasgow), one source's directional range, not a platform audit, but the thrust holds. A clip whose captions are mistimed or broken is a front door that won't open. Knowing the file means you can fix it instead of re-exporting and hoping.

How to hand-fix a broken SRT file

When an SRT won't load, it's almost never the video, it's a formatting typo in the text. Open the file in a plain-text editor (Notepad, TextEdit in plain mode, or VS Code, never Word, which can inject smart quotes and hidden characters). Then check these four things, in order:

1. Wrong millisecond separator. A file renamed from .vtt to .srt (or saved by a sloppy tool) often has periods where commas belong: 00:00:27.000 instead of 00:00:27,000. Find-and-replace the periods inside timestamps with commas. This is the single most common break.
2. A missing blank line between cues. If two cues run together with no empty line between them, the parser can merge them and everything after goes haywire. Make sure one blank line separates each cue. (Extra blank lines are usually harmless, most players tolerate them, but a missing one is a reliable way to break the file.)
3. A bad arrow. The separator is --> (two hyphens, a greater-than, with a space on each side). An em dash, a single hyphen, or a missing space will fail silently. Search for --> and confirm every cue uses it identically.
4. Wrong encoding. If captions show garbled characters, accents turning into é or boxes, the file was saved in the wrong encoding. Re-save as UTF-8 (every editor has this in Save As). For non-English shows especially, UTF-8 is the only safe choice.

A quick last check: open the file and scan that the sequence numbers count up cleanly and the timestamps move forward in time. A cue whose end time is before its start, or a later cue that starts earlier than the one before it, will throw some players off. Fix those by reading the timecodes, you now know exactly what each digit means.

The honest caveat: hand-fixing is great for a one-off rescue, but if your tool keeps producing broken files, the tool is the problem, not the file. A captioner that exports clean, correctly-timed cues every time is worth more than the best find-and-replace habit. The same goes upstream, getting good clips out in the first place depends on solid AI clip detection and a quick human pass to pick the best of the suggested clips before captions ever enter the picture.

FAQ

What does SRT stand for? SRT stands for SubRip Subtitle. It's named after SubRip, an early-2000s program that extracted subtitles from DVDs. The program is largely gone, but its plain-text format became the default way timed captions move between transcription tools, editors, and platforms like YouTube.

Can I open and edit an SRT file myself? Yes. An SRT is plain text, so any basic editor opens it, Notepad, TextEdit (in plain-text mode), or VS Code. Avoid Word, which can insert smart quotes and hidden formatting that breaks parsing. Edit the words or timecodes directly, keep the comma-before-milliseconds format, and save as UTF-8.

What's the difference between SRT and VTT? Both are timed-caption files with nearly identical structure. SRT uses a comma before milliseconds, carries no styling, and reads almost everywhere. WebVTT uses a period, can include styling and positioning, must start with a WEBVTT header, and is the web-native format for browser video players.

Do I need an SRT file for podcast clips on TikTok or Reels? Usually not directly. For clips posted natively, you burn the captions into the video, so the SRT is just the intermediate file your tool uses internally. You only hand-manage an SRT when you want a separate, toggleable caption track, most common when uploading long video to YouTube.

Why won't my SRT file load? Almost always a formatting typo: periods instead of commas in timestamps, a missing blank line between cues, a malformed --> arrow, or the wrong text encoding. Open it in a plain-text editor and check those four in order, the fix usually takes under two minutes.