The short answer
US universities produce audio description (AD) in four ways: ordering it from a managed service such as 3Play Media or through their video platform's marketplace, typing text descriptions directly into platform editors such as Panopto or Kaltura, recording narration in general-purpose video and audio tools, or writing and voicing it in a purpose-built desktop AD editor. At scale, most institutions end up with a hybrid: a vendor for high-stakes or long-form content, and an in-house team with a dedicated editor for the steady stream of course video.
The right mix depends on three numbers: how many minutes of video need description each year, what per-minute price you can negotiate, and how much of your content contains student data that shouldn't leave institutional systems. This guide walks through each.
Why the workflow question matters in 2026
Public universities fall under the Department of Justice's 2024 ADA Title II rule, which adopts WCAG 2.1 Level AA — including success criterion 1.2.5, audio description for prerecorded video. An Interim Final Rule published on April 20, 2026 moved the compliance dates to April 26, 2027 for public entities serving 50,000 people or more and April 26, 2028 for smaller ones, but left the standard unchanged.
The extra year changes the planning problem. Describing a backlog of a few hundred videos is a procurement exercise. Describing every new course video, every semester, is an operations problem — and that is where the choice of workflow and tooling decides whether the program is sustainable.
The four audio description workflow models
1. Managed description services
You upload video, a vendor's describers write the script, and you receive a finished track. 3Play Media, the most common vendor in US higher education, combines human describers with synthesized speech. Its public price list shows $9.00 per minute for standard English description with a synthesized voice, $14.00 for extended, and $18.00 and $35.00 for the voice-artist equivalents, with a standard five-business-day turnaround. Many universities have negotiated contracts: published institutional rates for standard description sit around $7–$8 per minute. Video platforms also resell vendor description; Kaltura customers typically order it through the REACH marketplace, and the finished track is attached to the video automatically.
Strengths: no staffing, predictable quality, platform integrations. Limits: cost scales linearly with minutes, turnaround is measured in days, and every file leaves your systems.
2. Text descriptions in the video platform
Panopto lets instructors type descriptions at timestamps in its editor or upload a description file; the player reads them aloud with a synthetic voice and pauses the video when needed. Kaltura's extended audio description works the same way: a time-coded VTT file voiced by the browser's speech synthesis, pausing playback as required.
Strengths: free with the platform, fast for short clips, faculty can do it themselves. Limits: no help finding gaps in dialogue, no review workflow, voice quality depends on the viewer's browser, and Kaltura's extended description is not supported in mobile playback. It works for a single lecture; it does not organize a program.
3. General-purpose editing tools
Many campus guides, such as the University at Albany's audio description resource, suggest planning and scripting descriptions, then recording them with a platform tool or a computer's built-in sound recorder. Media teams often extend this with Premiere Pro, Audition, or Resolve. Strengths: tools the team already owns. Limits: every gap is found by ear, timing is manual, and each revision means re-recording and re-mixing.
4. Purpose-built desktop audio description editors
Dedicated AD software puts the whole describer workflow in one application: detecting scenes and dialogue gaps, writing and timing descriptions against the picture, voicing them, reviewing, and exporting delivery files. It is the smallest category of the four, and the one designed for teams that describe video as ongoing work rather than as a one-off. Running on the desktop also means video can be processed on institution-managed machines instead of being uploaded to a third-party service.
What "large-scale" really requires
Scale changes the question from "how do we describe this video" to "how do we describe every video, consistently, every term." The programs that manage it share a few traits:
- Triage before production. Many lecture videos need little or no description because the instructor already narrates the visuals. Sorting content into no description, standard description, and extended description is the biggest cost lever available.
- One written standard. Usually the DCMP Description Key, extended with house rules for equations, charts, code, and slides — applied to in-house and vendor work alike.
- Separate roles. Describers write, reviewers approve, and a coordinator manages the queue and platform uploads.
- A record of what was done. Which videos were described, by whom, when, and against which standard. When a complaint or audit arrives, this record is your evidence of a good-faith program.
- Reusable delivery. Output formats that every campus platform accepts, so the same work can go to Kaltura, Panopto, Canvas, or a public site without rework.
Delivery formats: know what your platform accepts
Audio description reaches students in one of three forms, and your production tool should be able to produce the ones your platforms use. A mixed audio track contains the original sound plus narration and is selected from the player's audio menu; Kaltura, for example, accepts it as a manually uploaded audio file (AAC audio in an MP4 container), labeled clearly and uploaded alongside the original audio so viewers can switch between them. A text description track, usually WebVTT, is voiced by the player and is how Panopto and Kaltura's extended description work. An open-described version is a second copy of the video with description mixed in, the fallback for platforms with no track switching.
Human or high-quality recorded narration generally sounds better than browser speech synthesis, which varies by device. If your platform only accepts text tracks, the quality of the writing carries the whole experience.
Security and student privacy
Class recordings are not just media files. When a recording includes identifiable students asking questions, presenting, or taking part in discussion, those portions can be education records under FERPA, as university counsel offices such as UNC Charlotte's explain. Sharing them with a description vendor is usually handled under FERPA's school-official exception, which the Department of Education's guidance for service providers describes — with contracts, data controls, and vendor review to match.
That is manageable, but it adds procurement work and risk for every vendor in the chain. A desktop workflow where video stays on institution-managed machines keeps sensitive recordings inside your own environment. For seminar recordings, clinical training video, or research footage, that can decide which workflow is allowed at all. Check with your privacy office before you choose.
The cost math: vendor vs. in-house
At a negotiated $7 per minute, describing 100 hours of video costs $42,000; at the $9 list price, $54,000. Extended description roughly adds half again. Those figures repeat every year for new content. In-house production replaces that with staff time plus tooling, so its economics depend on how efficient each describer is: every minute saved finding gaps, retiming, and re-recording goes straight into capacity.
A realistic plan for most universities: triage first, send long-form and high-visibility content to a vendor, and build in-house capacity for short course video, where volume is highest and turnaround matters most.
Checklist: evaluating professional audio description software
- Gap and dialogue detection: does it find the pauses between speech and flag descriptions that overlap dialogue?
- Standard and extended description: can the same project produce both, since dense lecture content often needs extended?
- Timeline and script views: describers work differently; forcing one mode slows some of them down.
- Narration: built-in text-to-speech, human voice recording, or both, voiced inside the editor rather than in a separate tool.
- Export formats: a mixed audio track and a WebVTT description file at minimum, tested against your actual platforms.
- Local processing: can video stay on your machines, and what, if anything, is sent to external services?
- Review and audit trail: approval steps and a history of who changed what, for quality control and compliance records.
- Accessibility of the tool itself: ask for an Accessibility Conformance Report (VPAT). Blind and low-vision reviewers are among the best people to check description quality, and the software should work for them.
- Platform and IT fit: Windows and macOS support, signed installers, license management your IT team can deploy.
Where Synchrogen fits
Synchrogen is a desktop audio description editor for Windows and macOS, built for the fourth workflow above. It detects scenes and dialogue, shows the gaps on a timeline, and lets describers switch between timeline and script views while they write and time each description. Local text-to-speech voices the script inside the app, and finished work exports as a description track. Video is processed on the describer's machine rather than uploaded to a cloud service. The describer reviews and approves every line. Synchrogen is currently in closed beta, and university media teams can apply for a pilot. If you are building an in-house team, start with our guide to audio description training for higher education.
FAQ
What software do universities use for audio description?
Most US universities combine a managed service such as 3Play Media, often ordered through Kaltura REACH or similar integrations, with the text-description editors built into Panopto and Kaltura. Media teams that produce description in-house use general editing tools or purpose-built desktop AD editors.
How much does audio description cost for university video?
3Play Media's public list price is $9.00 per minute for standard synthesized-voice description and $14.00 for extended, with voice-artist narration at $18.00 and $35.00. Published university contract rates for standard description are around $7–$8 per minute.
Is a desktop editor more secure than a cloud service?
It reduces exposure when video is processed locally, because recordings that may contain student information do not have to be shared with another party. Security still depends on how your institution manages the machines and files, so involve your IT and privacy offices.
Can Panopto or Kaltura handle audio description on their own?
Both can play description, and both let you type or upload text descriptions that the player voices with speech synthesis. Neither is a full production environment: finding gaps, reviewing quality, and managing volume are left to you or to an ordered service.
Do all lecture videos need audio description?
No. Under WCAG, description is needed only for visual information that the main audio does not already convey. Lectures where the instructor narrates the slides may need little or none. See our WCAG audio description guide for the criteria.