File Tools. Local by design

Subtitle converter

SRT, VTT or just the words. Keep your captions moving.

Local processing · no upload

Your subtitles, ready for the next player.

or drop one file here · up to 2 MiB · 10,000 cues

Or paste subtitle text

Choose a file, paste subtitles or try the sample.

For standalone text captions only. Exports plain captions, not a full style-preserving subtitle project. No translation, speech recognition, subtitle extraction from video or timing repair. Output is UTF-8 without BOM; line endings are normalized to LF.

Convert standalone subtitles between SRT, VTT and plain text

Prepare plain captions for a web player, a video editor or your notes. Open SRT or WebVTT, check the timeline and export UTF-8. Every cue is validated before a download is created.

How to use this tool

  1. Choose an SRT or VTT file up to 2 MiB, paste subtitle text or try the sample.
  2. Check the timeline and source encoding. Errors identify a line and cue; overlapping captions are reported without changing their times.
  3. Choose SRT, WebVTT or TXT. Review and confirm any formatting or metadata removal, then create and download your copy.

A practical example

The SRT cue 00:00:01,250 --> 00:00:03,500 becomes 00:00:01.250 --> 00:00:03.500 beneath a WEBVTT header. The start, end and dialogue stay the same. TXT keeps dialogue and line breaks but omits the timing.

Frequently asked questions

Are styles and cue positions preserved?

No. This first version exports plain captions. Recognized formatting, speaker/language tags, inline timing, cue identifiers, positions, comments, styles and regions can be removed. The detected losses must be acknowledged before download. Unknown markup and live-stream timestamp maps are rejected.

Can I generate subtitles from a video?

No. This tool reads existing standalone SRT or VTT files. It does not transcribe, translate, burn in captions or extract embedded subtitle tracks.

Does it fix subtitle synchronization?

No. Cue times and overlaps are preserved. Invalid timestamps, reversed or zero durations, out-of-order start times and empty captions block conversion rather than being silently repaired.

Why do accented characters look wrong?

Check the source encoding. UTF-8, UTF-16 LE/BE, Windows-1252 and GBK are available. Legacy suggestions are uncertain; confirm the preview. All downloads use UTF-8 without BOM and LF line endings.

What are the limits?

One file or pasted text up to 2 MiB, 10,000 cues, 10,000 plain-text characters per cue, timestamps within 24 hours and output up to 4 MiB. The preview shows up to 100 cues; downloads include all validated cues.