AI audio description

Audio description for blind and low-vision viewers

Upload a video and AI writes timed descriptions of what is seen on screen: people, actions, places and on-screen text. Get a WebVTT descriptions track, an SRT and a script, ready for accessibility (WCAG, ADA, EAA).

Describe a video $0.30 per minute of video
Audio description generator opens soon.

Upload the video

MP4, MOV or WebM, up to 30 minutes

Video
The descriptions are written for the moments without speech where possible.
$0.30 per minute of video

How it works

1

Upload the video

A lesson, a product video, a short film or an ad.

2

AI watches it

It describes what matters to follow the story, between the spoken words.

3

Download the track

WebVTT (kind "descriptions") for web players, SRT, and a text script to check or record.

Made for

Schools and universities

Accessible lecture and course videos.

Companies and public bodies

Meet accessibility rules for training, marketing and public information.

Creators

Reach blind and low-vision viewers on your channel.

Pricing

Audio description generator

Please try before you buy, to make sure it meets your needs

Free 0$
  • Use the free credit of your account
  • Same quality, no watermark
Pro 0.30 $/min
  • As many as you like
  • Money back if a job fails

Starter

$15 credit
About 50 minutes of this tool

Pro

$99 credit
About 330 minutes of this tool

Business

$299 credit
About 996 minutes of this tool

The credit pays for the tools priced in $ per minute, like this one. Subtitle minutes are sold on the pricing page.

Secure checkout with PayPal or card
Secure checkout by PayPal: PayPal, Visa, Mastercard, Amex 320,000+ accounts · 150,000+ videos processed
  • Payment not showing? Email your PayPal receipt to and we add it to your account right away.
  • Help when you need it. Questions about a job, a file or a payment? Contact support, we answer every message.

Questions

What is audio description?

A narration of the important things that are seen but not said: who appears, what they do, where they are, text on screen. Blind and low-vision viewers hear it between the dialogue.

What do I get?

A WebVTT file with kind "descriptions" (web players and screen readers read it), the same as SRT, and a plain text script.

How do I get it as a voice?

Put the SRT into Subtitles to speech to get a timed voice track, then mix it under your video.

Is it perfect?

AI can miss details or name things wrongly. Read the script before you publish, especially for legal or medical content. For checked work, ask for our human-reviewed service.

How much does it cost?

See the price above, by the minute of video. Taken from your credit when the job starts and given back if it fails. Free credit works here.

What happens to my video?

It is processed by our AI partner Replicate (United States, Google Gemini model) only to write the descriptions. See the Privacy Policy.

Your privacy comes first. Files are processed by our AI partner Replicate (United States) only to make your result. See the Privacy Policy.