SRT, VTT, ASS, SSA, SUB, SBV and plain text — converted in your browser. The file never leaves your device, because there is no server to send it to.
MicroDVD .sub files count frames, not seconds — this has to match the video.
Seven formats, every direction. Drop a file in and the format is detected from what is inside it, not from the extension — which matters more than it sounds, because .sub alone is used by three unrelated formats.
SRT is the plainest subtitle format there is: a number, a start and end time, the lines of text, a blank line, repeat. It came out of a 1990s DVD-ripping program called SubRip and has outlived every format designed to replace it, precisely because there is almost nothing in it to go wrong.
Where you meet it: It is what almost every subtitle site hands you, what VLC and MPV expect next to a video file, and what Netflix, YouTube and Vimeo all accept on upload.
What it costs you: No styling worth the name. Position, font, colour and karaoke timing simply do not exist in SRT — players allow italics and a few tags by convention, not by specification, so two players can disagree about the same file.
WebVTT is SRT rebuilt for the web: the same cue structure, a WEBVTT header line, a dot instead of a comma before the milliseconds, and optional positioning settings after the timestamp. It is the only subtitle format HTML5 video natively understands.
Where you meet it: Any <track> element on a web page, HLS and DASH streams, and most browser-based players. If subtitles have to work in a browser without a plugin, they end up as VTT.
What it costs you: Desktop players are less consistent with VTT than with SRT, and a few older TVs and set-top boxes ignore it entirely. Its positioning settings are also the first thing lost when the file is converted to anything else.
ASS is a typesetting language that happens to carry subtitles. Named styles, exact positioning, rotation, fades, karaoke timing and vector drawing all live in the file, split across a [Script Info], a styles and an [Events] section.
Where you meet it: Fansubbed anime, almost universally, and any release where the signs on screen are translated in place. Aegisub writes it; mpv, VLC and MKV containers read it.
What it costs you: Everything that makes it powerful is what breaks elsewhere. Upload an ASS file to a platform that wants SRT and the styling is stripped at best, or the file is rejected outright. It is also the one format here that is genuinely hard to hand-edit.
SSA is ASS's predecessor — version 4.00 of the same format, before the "Advanced" extensions were added in v4.00+. The sections and the Dialogue lines look nearly identical; the first field is Marked rather than Layer, and the styles block has fewer columns.
Where you meet it: Older fansub releases and archives from the early 2000s, and the occasional piece of software that never moved on to v4.00+.
What it costs you: It has ASS's compatibility problems without ASS's capabilities. Unless something specifically demands v4.00, there is no reason to author in SSA today.
MicroDVD is the odd one out: it counts frames, not time. Each line is {start}{end} followed by the text, with a pipe between display lines. There is no timestamp anywhere in the file.
Where you meet it: DivX-era AVI files, older hardware players, and subtitle archives that have not been touched since. The extension .sub is also used by SubViewer and by VobSub's binary format, which are different things entirely.
What it costs you: A frame-based file is only correct at the frame rate it was made for. Play a 23.976 fps subtitle against a 25 fps video and it drifts about four percent — roughly three minutes late by the end of a feature film.
SBV is YouTube's own caption format, and about as simple as SRT with one line less: no sequence number, and the two timestamps separated by a comma rather than an arrow.
Where you meet it: Downloading captions out of YouTube Studio, mostly. It is the default export there, which is why people end up holding one.
What it costs you: Almost nothing outside YouTube reads it. No styling, no positioning, and desktop players will generally refuse the file until it is converted.
Plain text is the subtitle with the subtitle part removed: the spoken lines, in order, with no timing at all.
Where you meet it: Wherever the words matter and the timing does not — reading a transcript, feeding a translator, counting words, or pasting a script into a document.
What it costs you: The timing is gone and cannot be recovered. Converting back out of plain text means inventing timings from scratch, which is a starting point for a human to correct, not a finished subtitle.
No. The conversion runs in your browser — the file is read with JavaScript, converted in memory and handed back as a download. Nothing is transmitted, so there is nothing to delete afterwards. You can disconnect from the internet after the page loads and it still works.
SRT, WebVTT, ASS, SSA, MicroDVD .sub, YouTube SBV and plain text — all seven in both directions, which is 42 conversions in total. The source format is detected from the file's contents rather than its extension.
Only when MicroDVD is involved. That format stores frame numbers rather than timestamps, so converting it requires knowing the video's frame rate. Get it wrong and the subtitles drift steadily — 23.976 against 25 fps is about four percent, which is three minutes by the end of a film. If the file declares its own rate on a {1}{1} line, that is used instead.
Styling, when the target format has none: converting ASS or SSA to SRT drops positioning, fonts, colours and karaoke timing, because those things cannot be written in SRT. Converting to plain text drops the timing entirely. Both are stated on screen after the conversion rather than left for you to discover.
Yes, but be clear about what you get: each line becomes a cue of a fixed length, laid end to end. The words are right and the timing is invented. Use the subtitle sync tool afterwards to line it up against the video.