Subtitles & Captions
TXT to SRT: How to Add Real Subtitle Timings to a Transcript
A .txt transcript becomes an SRT only when every caption has a number, a start and end time, and its text. Renaming the file does not add times. Split the text into short captions, then set each start and end by typing it or by marking it while your video plays, check that no captions overlap or flash too fast, and save the file as plain UTF-8. Estimated timings are a starting point only, because they are not synchronized to speech.
By Tubelexity / Updated
Open TXT to SRTWhat an SRT caption contains
Each SRT caption is a sequence number, a timing line with start and end times to the millisecond, one or more lines of text, and an empty line before the next caption. YouTube lists SubRip (.srt) as a basic caption format and says the file must be plain UTF-8 without styling markup.
For a two-line transcript, “Hola, ¿qué tal?” followed by an empty line and “Press <Enter> to start.”, timed at 0–1.5 s and 1.5–4 s, the tool writes exactly: 1 / 00:00:00,000 --> 00:00:01,500 / Hola, ¿qué tal? / (empty line) / 2 / 00:00:01,500 --> 00:00:04,000 / Press <Enter> to start. The accented characters are UTF-8, and the angle brackets are kept with a warning because some players read them as styling.
Three ways to add times
Only the first two produce timings that follow the speech. Use the third to get a draft on screen quickly, then correct it.
- With “start the next caption at the same time” on, pressing E also starts the next caption, so continuous speech needs one key per caption.
- Two captions estimated over 10 seconds share the time by length: the longer caption gets more of it, and the total is exact.
| Method | How it works | Accuracy |
|---|---|---|
| Mark against the video | Press S at the first word and E at the last while it plays | As good as your timing; easiest to correct |
| Type times | Enter 00:01:02,500, 1:02.5 or 62.5 | Exact to the millisecond you enter |
| Estimate | Spread captions over a total duration by text length | Not synchronized to speech; labelled Estimated |
Checks before you export
A caption without a start or end, or with an end before its start, blocks the export. Nothing is deleted automatically: fix the time or delete the caption yourself. Overlaps, captions out of order, very short or very long captions, and reading speed above 21 characters per second are warnings. Those numbers are reference ranges, not platform rules.
Two characters can break an SRT file inside a caption: an empty line (it ends the caption early) and the arrow “-->” (it looks like a timing line). The tool asks you to remove both.
What this does not do
TXT to SRT does not transcribe, translate or listen to audio. It does not align text to speech automatically. Your text, times and video stay in the browser; a draft is kept in browser storage only if you turn that option on.
Recommended workflow
- Open TXT to SRT. Paste the transcript, or choose Upload a UTF-8 .txt file.
- Under Split into captions by, choose Empty lines, Each line or Sentences, then select Create caption blocks.
- Optionally open your video. Click the timing panel, press K to play, S to set a caption's start and E to set its end and move on.
- Or type each Start and End, such as 00:01:02,500 or 62.5.
- Fix every caption marked as needing a time, review the warnings, then choose Copy SRT or Download SRT.
Frequently asked questions
How do I convert a TXT file to SRT?
Split the text into captions, give each caption a start and end time, and export. A transcript has no times, so renaming .txt to .srt does not work.
Can I make an SRT file without timestamps?
No. Every SRT caption needs a start and end time. You can estimate them first, but they are not synchronized to speech until you check them.
What encoding should an SRT file use for YouTube?
YouTube asks for plain UTF-8. TXT to SRT writes UTF-8 without a byte-order mark.
How long should each subtitle stay on screen?
Long enough to read: the tool warns below 0.8 seconds, above 7 seconds and above 21 characters per second. These are reference ranges; audiences and languages differ.
