Captions

Captions in the language it was spoken.

Upload a video or its audio and download subtitle files ready for your player, platform or editor — timed to the speech. Need them checked by a person? We do that too.

News report — isiNdebelecaptions by Bhala
Captions unedited — exactly as our system wrote them.Video: Voice of America (public domain)

What you get

SRT and VTT, ready to use

The two formats every video player, social platform and editing tool accepts.

Timed to the speech

Every line starts and ends with what is being said, so captions never run ahead or lag behind.

Readable lines

Lines break at natural pauses and stay short enough to read comfortably on screen.

Long videos

Up to an hour per file, in every common audio and video format.

Check before you publish

Play the recording with the transcript beside it and see the less certain words underlined.

Reviewed by a person

For broadcast, education or public-sector work, have captions checked by a reviewer before delivery.

Two ways to buy

Machine captions

Per minute, billed by the second

Self-serve from your dashboard. Upload, review, download.

  • SRT, VTT, TXT and JSON
  • Timed to the speech
  • Detects the language
  • Free credit to start

Reviewed captions

Quoted per project

Machine captions, checked and corrected by a person before delivery.

  • Everything in machine captions
  • Human review before delivery
  • For broadcast, education and public sector
  • Volume pricing

How it works

  1. 1Upload the video or its audio
  2. 2Review the transcript alongside the playback
  3. 3Download SRT or VTT — or send it for review

Who uses it

Video teams & publishers

Caption every video you post, in the language your audience speaks.

Education & training

Accessible lectures and training videos in local languages.

Government & public services

Captioned public information, briefings and campaigns.

Agencies & production houses

Deliver captioned work to clients without a manual transcription step.

Languages: isiZulu, isiXhosa, Sepedi, Sesotho, Setswana, isiNdebele (South Africa), isiNdebele (Zimbabwe), Kiswahili, Luganda, Lingála.

Questions

Can I upload video files?
Yes — MP4, WebM and other common formats. The audio is taken from the video; up to 200 MB and 60 minutes per file.
How are machine captions priced?
The same as transcription: priced per minute of audio and billed by the second, from prepaid credits.
How do reviewed captions work?
A person checks and corrects the captions before they are delivered. Pricing and turnaround depend on the volume and the language — talk to us for a quote.
Do you caption live streams?
Not yet. Captions are made from recorded audio and video.