No account required

Generate your video's subtitles automatically

Upload the video and get an SRT or VTT file with the timings calculated. You can edit it before downloading.

Drag your file in here

MP3, WAV, M4A, MP4, MOV… and any other audio or video, whatever the size

  • SRT and VTT, with the timings already calculated
  • Speaker identification on each line
  • Edit the text before downloading
  • MP4
  • MOV
  • MKV
  • WEBM
  • MP3
  • WAV
  • M4A

About this tool

Creating subtitles manually is the slowest part of any video production: play, pause, type, and then just as much time again adjusting the timings. This tool handles the first pass in about three minutes per hour of video, leaving for human review only what genuinely needs it: a proper name, a technical term.

The result isn't a block of text: the lines are split correctly, the timings are synchronised to the millisecond with the audio, and when there are several speakers, each line indicates who's speaking.

Features

  • Timings to the millisecond, calculated from the audio, not estimated.
  • Lines split to a readable length, instead of full paragraphs.
  • Speaker identification when the recording contains several voices.
  • Over 90 languages, with a translated track available alongside the original.

How it works

  1. 1

    Upload the video or audio

    MP4, MOV, MKV, WEBM, MP3, WAV or M4A, up to 50 GB.

  2. 2

    Review the result

    Play the file with the synchronised text and correct anything necessary.

  3. 3

    Download the subtitles

    SRT for editors and platforms; VTT for browser players.

Use cases

Social media video

Most viewers watch video without sound; subtitles are essential.

Accessibility

They make content accessible to people with hearing impairments.

Courses and training

A subtitled class can be searched by content, in addition to serving as a transcript.

In detail

How much text fits in a subtitle

The usual convention is two lines of about forty characters, visible on screen long enough to read comfortably: between one and seven seconds, never under a second. This isn't an arbitrary rule; it's what became standard in television and cinema over decades, and players assume it's being followed.

Segments are split according to that criterion, not sentence by sentence. If you edit the text afterwards to fix a name, the cut point doesn't change: what changes is what's read, not when each subtitle appears and disappears.

SRT or VTT? It depends on the destination

SRT when the destination is an editor, a desktop player or a platform's upload field. VTT when the video is embedded on your own website, since the browser's player only supports that format.

Both can be downloaded from the same screen without transcribing again. If you later need to convert between the two, the relevant converters are available with no account required.

Frequently asked questions

Need to transcribe a video?

Upload it and get the text with timestamps, each voice identified and a summary. No account needed to start.