How each subtitle cue is built
Every caption segment becomes one numbered SRT cue: a sequence number, a time range in HH:MM:SS,mmm format (SRT uses a comma before the milliseconds), and the caption text, followed by a blank line.
Start times come straight from the caption track. The end time is the segment’s own duration, cut short if the next caption starts earlier, so cues never overlap on screen. Very short cues are held for at least half a second so they stay readable.
Sample SRT output (original example text, not from a real video)
1 00:00:00,000 --> 00:00:04,000 First, write down the question you want the video to answer. 2 00:00:04,000 --> 00:00:08,500 Next, search the transcript for a distinctive phrase. 3 00:00:08,500 --> 00:00:13,500 Finally, check the matching moment in the source video.
Using the SRT file
- Video editors: import it as a caption or subtitle track in tools such as Premiere Pro, DaVinci Resolve, or Final Cut (via a captions import), then restyle and retime there.
- Media players: VLC and most desktop players load a .srt automatically when it has the same file name as the video, or via the subtitle menu.
- Re-uploading: YouTube and many course platforms accept SRT uploads, which is useful when you have edited the text and want to replace auto-generated captions on your own video.
- Translation: SRT is the format most subtitle translators and translation tools expect, because the timing survives the round trip.
Before you publish subtitles
- Auto-generated captions can be several words long per cue or split mid-sentence. For burned-in or broadcast subtitles, re-segment long cues in your editor.
- Check names, numbers, and technical terms; auto captions commonly get these wrong.
- The file is UTF-8 encoded. Older players that expect another encoding may show garbled accents; modern editors handle UTF-8 correctly.
- Only publish subtitles for videos you own or have permission to caption.