UtiliKit
Subtitle Generator · Beta

Auto Subtitle Generator

Add captions to a video, or turn any recording into text. Speech recognition runs in your browser, so the file never leaves your device.

  • No sign-up
  • 99 languages
  • SRT, VTT and MP4
  • Nothing is uploaded

Drop a video or audio file here

or click to choose one. MP4, MOV, WebM, MKV, MP3, WAV, M4A and more.

The speech model downloads once and is then kept by your browser. A 10 minute video takes roughly 2 to 5 minutes on a recent laptop.

Subtitles without uploading your video

Most caption tools send your video to a server and bill by the minute. This one downloads a speech model into your browser and does the work on your own computer. That keeps private recordings private, and it means there is no queue and no minute counter.

Automatic transcription

Speech is recognised in 99 languages, with the language detected for you. Each word gets its own timing, which is what makes word-by-word captions possible.

Caption styles for short video

Bold captions that show a few words at a time and light up the word being spoken, the look used on Reels, Shorts and TikTok. Classic and boxed subtitles are there too.

Vertical, square or original

Turn a wide video into a 9:16 clip by cropping, or keep the whole picture over a blurred background. Captions are placed for the new shape.

SRT, VTT and text

Download subtitle files for YouTube, Vimeo, LinkedIn or your editing software, or copy the transcript as plain text for show notes and articles.

Fix it before you export

Every caption can be edited. Correct a name, remove a filler word or adjust the timing, and the preview updates straight away.

Translate to English

Speak in Spanish, German, Lithuanian or any other supported language and get English captions in one step.

Free and Premium

Transcribing, editing and subtitle files are free. Premium removes the small mark from exported videos, lifts the length limit and opens the extra caption styles.

Free Premium
Transcription and caption editing First 30 minutes of each file Whole file, up to 3 hours
SRT, VTT and text files Unlimited Unlimited
Video with burned-in captions Unlimited, with a small utilikit.eu mark Unlimited, no mark
Caption styles Classic, Boxed, Bold Pop All styles, including Karaoke and Spotlight
Vertical, square and portrait frames Included Included

See Premium

How to add subtitles to a video

  1. Drop your video or audio file on the box above. Leave the language on automatic or pick it yourself.
  2. Wait while the speech is turned into text. The first run also downloads the speech model.
  3. Read through the captions and fix anything that was misheard.
  4. Pick a style, set how many words appear at once, and choose the frame shape.
  5. Download the SRT or VTT file, or export the video with the captions burned in.

Good to know

Accuracy depends on the recording. Clear speech with little background noise gives the best result; music, crosstalk and strong accents cause more mistakes, so always read the captions before you publish. The larger model is slower but noticeably better with names, accents and languages other than English. Video export works in Chrome, Edge and other current browsers on a computer; phones can transcribe, but long videos need a lot of memory.

Credits and licences

Speech recognition uses Whisper by OpenAI (MIT licence) through Transformers.js (Apache-2.0). Video reading and writing uses Mediabunny (MPL-2.0). Caption fonts are Montserrat, Poppins, Anton and Bangers under the SIL Open Font Licence (details).

Related tools: Video Compressor, Video Cropper, Photo Editor.

Frequently Asked Questions

Is the subtitle generator free?

Yes. Transcribing, editing captions and downloading SRT, VTT or text files is free with no account. Free use covers the first 30 minutes of each file, and exported videos carry a small utilikit.eu mark. Premium removes the mark, handles whole files up to 3 hours and adds more caption styles.

Is my video uploaded?

No. The speech model is downloaded to your browser and runs on your own device. The video or audio file is read locally and is never sent to a server.

Which languages are supported?

99 languages, including English, Spanish, French, German, Portuguese, Italian, Dutch, Polish, Lithuanian, Ukrainian, Russian, Turkish, Arabic, Hindi, Japanese, Korean and Chinese. The language is detected automatically, or you can choose it. Accuracy is best for widely spoken languages.

How long does it take?

On a recent laptop the Balanced setting works through speech about two to five times faster than real time, so a 10 minute video takes 2 to 5 minutes. The first run also downloads the speech model, which is kept for next time. Older computers and phones are slower.

Which file types can I use?

MP4, MOV, WebM and MKV video, and MP3, WAV, M4A, AAC, OGG, Opus and FLAC audio. Exported videos are MP4 with H.264 where the browser supports it, otherwise WebM.

Can I get an SRT file without exporting a video?

Yes. As soon as the transcript is ready you can download SRT, VTT or plain text, or copy the text. You only need the video export if you want the captions burned into the picture.

Can I translate subtitles?

You can turn speech in any supported language into English captions by choosing English under "Captions in". Translation into other languages is not available yet.

Why are some words wrong?

Automatic speech recognition makes mistakes with names, unclear speech, background music and overlapping voices. Click any caption to correct it, and try the Most accurate setting for difficult recordings.