Subtitle Converter

The three formats look similar but are not interchangeable, and using the wrong one is the usual reason captions load blank.

SRT (SubRip) is the most widely supported subtitle format — media players, video editors, YouTube uploads, and most social platforms all take it. A cue is a number, a timecode line, and the text:

1
00:00:01,000 --> 00:00:04,000
Welcome back to the show.

Note the comma before the milliseconds. That comma is the SRT signature.

VTT (WebVTT) is the format the web uses. It is what an HTML5 <track> element inside a <video> tag expects, so it is what you need for captions on a website or in most JavaScript video players. It must start with a WEBVTT header, the cue numbers are optional, and the timecode uses a period, not a comma:

WEBVTT

00:00:01.000 --> 00:00:04.000
Welcome back to the show.

VTT can also carry styling, positioning, and cue settings that SRT cannot — those are dropped when you convert down to SRT, since SRT has nowhere to put them.

TXT is not a subtitle format at all. It is the plain transcript underneath: every spoken line with the numbers and timecodes removed, ready to paste into a document, a blog post, or anywhere you want the words without the timing.

Converting between SRT and VTT preserves your timing exactly. Every cue keeps its start and end time down to the millisecond — the tool only rewrites the surrounding syntax. Going to VTT, it swaps each comma for a period and adds the WEBVTT header; going to SRT, it does the reverse and renumbers the cues sequentially. The words on screen appear at precisely the same moments as before.

Choosing TXT is the one direction that drops timing on purpose. The timecodes and index numbers come off and you are left with just the dialogue as continuous readable text. That is what you want for a transcript, show notes, translation, or feeding the words into another tool — but it is a one-way trip. Once the timestamps are gone, they are gone, so keep your original SRT or VTT if you might still need synced captions later.

Subtitle files should be UTF-8, and this converter keeps them that way. UTF-8 is what carries accented letters, non-Latin scripts, and punctuation like curly quotes and em dashes correctly. The classic bug — names showing up as é or ’ — is almost always a subtitle file saved in some legacy regional encoding instead of UTF-8, then read by a player that assumed UTF-8.

Because the text is read and rewritten in your browser as Unicode, the characters you paste in are the characters you get out, intact. The tool does not add a byte-order mark (BOM) to the output, which is the safer default: a stray BOM at the top of a file can push some strict players into misreading the very first cue or the WEBVTT header. If your source file already displays correctly here in the result pane, it will export correctly.

Subtitle conversion does not need a server. It is string manipulation — reformatting timecodes, adding or removing a header, stripping numbers — so there is no reason to upload your file anywhere to do it.

Running it locally has real upside beyond privacy. It is instant, with no upload-and-wait, so a three-hour subtitle track converts as fast as a one-minute clip. There is no practical size limit, because you are bounded by your own machine, not an upload cap. And your captions — which for an unreleased film, a private recording, or a client's video can be genuinely sensitive — never leave your device. You can confirm all of this in the network tab: no request fires when you convert.