Platform reference · D2L Brightspace · LTI 1.3 · AODA + ADA Title II

D2L Brightspace captions: Insert Stuff, D2L Capture, and the AODA + Title II retrofit

D2L Brightspace is the dominant LMS in Canadian higher education — by a meaningful margin in Ontario, where almost every public university and college is on a Brightspace tenant — and increasing in US K-12 and corporate L&D. Captioning a Brightspace course catalogue runs through three video surfaces: video uploaded through the Insert Stuff dialog with a sidecar VTT or SRT, D2L Capture's lecture-recording pipeline (formerly D2L Capture Central / Capture Live), and external-tool video (Panopto, Kaltura, YouTube, Vimeo, Zoom recordings) embedded through an LTI tool. The compliance pressure is layered: AODA's IASR binds Ontario tenants to WCAG 2.0 AA with a three-year compliance-reporting cycle (next major large-organisation filing window: 2026), and ADA Title II's 2026-04-24 deadline binds US tenants to WCAG 2.1 AA. The back-catalogue retrofit is urgent at every Brightspace tenant in scope.

TL;DR

A Brightspace captioning workflow has three surfaces. (1) Insert Stuff video — instructors upload an MP4 through the Insert Stuff dialog and attach a sidecar VTT or SRT caption file; the Brightspace video player reads the sidecar through HTML5 track. (2) D2L Capture — D2L's lecture-recording solution (Capture Central / Capture Live), with lecture recordings, slide-sync, and an attachable caption track. (3) External video via LTI — most large-volume video lives in Panopto, Kaltura, YouTube, or Zoom recordings embedded through LTI. AODA's IASR section 14 (O. Reg. 191/11) binds Ontario tenants to WCAG 2.0 AA with three-year compliance reporting; ADA Title II binds US tenants to WCAG 2.1 AA at the 2026-04-24 deadline. The retrofit pattern: inventory the catalogue → identify which surface owns each asset → re-caption the high-stakes content first → publish glossary-biased captions back to each surface → log the asset register for the next ACR filing window. The proper-noun mangling pattern that catches institutions in audit is the same in both regimes.

Brightspace's market position — why this is concentrated in Canada and selected US states

D2L is headquartered in Kitchener, Ontario, and Brightspace is the default LMS at most of Ontario's publicly assisted universities and colleges, plus the equivalent default in Alberta, much of British Columbia, and a meaningful fraction of Quebec's English-language post-secondary system. In the US, Brightspace is concentrated in K-12 districts (where it competes with PowerSchool's Schoology), select state university systems (notably parts of the State University of New York and the University System of Georgia), and the corporate-L&D mid-market.

The compliance posture follows the geography. Canadian tenants face AODA in Ontario, the Accessible BC Act in British Columbia, the Accessibility Act in Manitoba, the Accessibility Act in Nova Scotia, and the Accessible Canada Act for federally regulated organisations. US tenants face ADA Title II at the 2026-04-24 deadline plus Section 504 for federal-financial-assistance recipients and Section 508 for federal-contractor flow-down. Cross-border SaaS companies running corporate-L&D content on Brightspace face all of the above plus EAA in the EU.

Surface 1 — Insert Stuff video with sidecar captions

The most common Brightspace captioning surface for instructor-created content is the Insert Stuff dialog inside the HTML editor. The flow:

Open the Brightspace HTML editor (in Content, Discussions, Quizzes, or any other surface that uses it).
Click Insert Stuff → Add Video Note (for browser-recorded short clips) or Upload a File (for an existing MP4).
For uploaded video, the dialog accepts a separate caption-track file in WebVTT or SRT format. Brightspace's player serves the captions through the HTML5 track element with a CC toggle in the player UI.
Save. The video and its caption track render inline in the Content topic, discussion thread, or quiz item.

The Insert Stuff surface has three operational realities relevant to a captioning retrofit:

Video Notes are auto-captioned (recent versions). Brightspace's Video Note feature — the in-browser short-form recorder — runs auto-captioning by default in current Brightspace tenants. The auto-captions land in the same 80-90% substantive-accuracy band as every other generic auto-captioner. For non-trivial proper-noun content the auto-caption is a draft, not a finished caption track.
Sidecar caption files don't auto-attach across course copies. Like Canvas, Brightspace course-copy moves the video and the caption file but the caption-track association in the HTML editor often has to be reattached. This is the most common point of caption detachment between terms.
Caption files are downloadable. The sidecar file is served from the LMS file system and can be downloaded by anyone with read access to the embedding page. This matters for any institution that puts student names, accommodation notes, or unredacted feedback into caption text — don't.

For format choice, see the SRT and VTT reference pages — Brightspace's HTML5 player handles both.

Surface 2 — D2L Capture (lecture recording)

D2L Capture is D2L's lecture-recording solution. Capture Central is the studio-recording variant (an instructor records into a Capture session, optionally with slide sync and webcam picture-in-picture); Capture Live is the streaming variant for live lecture sessions; both produce a video asset in the Capture library that can be embedded into Brightspace courses.

The Capture caption surface:

Auto-transcript with editable correction. Capture runs speech-to-text on session recordings and produces a per-cue caption track aligned to the timeline. The instructor or accessibility staff can edit the cue text and timing.
Caption upload and replace. Vendor-supplied SRT or VTT files can replace the auto-transcript wholesale. This is the supported workflow for re-captioning the back-catalogue with glossary-biased output.
Multi-language tracks. Capture supports multiple caption tracks per recording. The student-facing player exposes a language selector when more than one track is present.
Captions follow embeds. When a Capture recording is embedded in a Brightspace course (via the Capture LTI insert), the caption track follows the recording. This is what makes Capture the right surface for any video that recurs across multiple courses or terms.

The Capture caption editor is functional for line-by-line correction but doesn't scale to a back-catalogue retrofit. For volume work, the operational pattern is to bulk-export auto-captions out of Capture, run them through a glossary-biased re-captioning pass, and bulk-replace.

Many institutions have a parallel deployment of Panopto or Kaltura alongside D2L Capture — Panopto for high-volume lecture-capture in classrooms with appliance hardware, Capture for instructor desk-recorded short-form. The catalogue audit has to look at all of them.

Surface 3 — External video through LTI

The largest video catalogues in higher education don't live in Brightspace; they live in lecture-capture and video-hosting platforms embedded through LTI 1.3:

Panopto. Often the dominant lecture-capture platform at large research universities, integrated through the Panopto LTI tool which surfaces a course folder of recorded videos. Captions are uploaded inside Panopto and follow the video into Brightspace.
Kaltura. The Kaltura Video Package for Brightspace is the second most common LTI integration. Captions are managed in Kaltura's My Media or KMC.
Zoom recordings. Brightspace's Zoom integration ingests cloud recordings and surfaces them through a course tile. The auto-transcript is Zoom's; the captioning workflow is Zoom's; the auto-transcript failure mode is Zoom's.
YouTube. Embedded through the Insert Stuff Insert YouTube Video flow. Captions are the channel owner's responsibility; YouTube auto-captions are not sufficient for ADA Title II or AODA SC 1.2.2 substantive accuracy.
Vimeo. See Vimeo captions for training videos; up to five caption formats per video, embedded through Insert Stuff.
Microsoft Stream. See Microsoft Stream captions; common at corporate-L&D Brightspace tenants whose IT is on M365.
Loom. Async-video integration through Loom Education or Loom Business; see Loom captions.

The catalogue-inventory step has to look across all surfaces. AODA's compliance attestation and Title II's substantive-accuracy bar both apply to whatever the student sees through the LMS, regardless of the hosting platform.

The AODA compliance-reporting cycle, applied to a Brightspace tenant

AODA's IASR (O. Reg. 191/11) section 86 requires designated public-sector organisations and large organisations (50+ employees in Ontario) to file Accessibility Compliance Reports with the Ministry of Seniors and Accessibility on a recurring schedule — historically every three years for large organisations, with the most recent filing windows in 2014, 2017, 2020, 2023, and the next major large-organisation filing window in 2026.

The report attests in writing to compliance with each part of the IASR. Section 14's WCAG 2.0 AA bar is one of the items — and the Accessibility Compliance Authority's sampling pattern, when it follows up on a filing, is the same one a Title II OCR investigation runs:

Open the institution's public-facing portion of the website or LMS.
Find the training-video catalogue (often a public course catalogue, careers page, or open-access programme).
Watch a slice with captions on.
Read the caption track against the audio.
If the proper nouns the speaker is teaching are mangled, flag the attestation as out of conformance.

The "every three years" rhythm is what AODA enforcement uses. Institutions that filed "fully compliant" three years ago and now have a Brightspace catalogue with auto-captioned content full of mangled programme names, regulatory citations, and Indigenous community names are the obvious investigation targets in the 2026 cycle.

The proper-noun failure mode in Canadian Brightspace content

Operators who watch the AODA sampling pattern in Brightspace tenants report consistent surfaces:

Provincial regulatory citations. "PHIPA" → "pip-fa", "MFIPPA" → splintered, "OHSA" → letter-by-letter, "PIPEDA" → "pipita". Health-sector and public-sector training is dense with these.
Ministry programme names. "OHIP" → "owe-hip", "WSIB" → spelled out, "ServiceOntario" → "service Ontario", "OSAP" → "oh-sap".
Indigenous community and language names. "Anishinaabe", "Haudenosaunee", "Mississaugas of the Credit First Nation", "Nishnawbe Aski Nation", "Ininiw" — the words that institutions most need to get right in cultural-competency and Indigenous-studies modules are the ones generic STT has the least training data for. Mangling these in a public-facing module is exactly the failure that draws complaint and inspector attention.
French terms in bilingual training. Brightspace tenants in Ontario, New Brunswick, and parts of Manitoba carry French-language content; "Université de Hearst", "L'Université de l'Ontario français", "Conseil scolaire de district catholique du Centre-Sud" don't survive a generic Whisper pass without bias.
Healthcare drug, procedure, and instrument names. See the same surface as medical training video captions documents.

The shared pattern: the words a Canadian post-secondary student or workforce trainee must learn to do their job are exactly the words generic STT has never seen.

The back-catalogue retrofit pattern

For an institution sitting on years of un-captioned or auto-captioned Brightspace video, the retrofit runs in five phases:

Inventory. Use Brightspace's Valence API plus the Capture API plus the LTI-embedded platform APIs (Panopto, Kaltura, Zoom, Vimeo) to build a flat catalogue. Most institutions discover that 60–80% of the catalogue lives outside Insert Stuff.
Triage. Required courses first, regulated-content modules first within those (compliance training, healthcare procedure videos, anything that's audit-bait under PHIPA / HIPAA / Section 504), public-facing welcome video first within marketing.
Re-caption. Replace mangled or absent captions with glossary-biased output. The institutional glossary is built once — programme names, course names, faculty names, regulatory citations, drug and procedure names if you have a healthcare programme, SDK symbols if you have a CS programme, the institution's acronym handbook, the Indigenous community name register, the bilingual French terms — and applies to every retrofit asset.
Publish. Sidecar SRT/VTT for Insert Stuff; replace caption track in D2L Capture; upload through the platform API for Panopto/Kaltura/Vimeo/Zoom; channel-owner action for YouTube.
Log for the ACR filing. Maintain an asset register: video URL, surface, caption file, caption source, reviewer, review date, glossary version. This is the documentation that backs the ACR attestation in the 2026 filing window.

See pricing

Brightspace accessibility-checker behaviour

Brightspace ships with an accessibility checker built into the HTML editor that flags video accessibility issues at content-edit time. The checker reports:

Video without captions. A video embed without an attached caption track produces a flag at save time.
Video with auto-caption-only track. Some checker configurations and accessibility-plug-ins surface auto-caption-only as a partial fail, prompting a human-review action.
External-platform videos. The checker can't reach inside the LTI-embedded platform to verify captions; institutional compliance posture has to assume LTI-embedded video may need separate audit.

The checker is a useful gating signal at content-creation time but doesn't substitute for the catalogue-wide audit. Most legally exposed videos are months or years old.

Where glossary-biased captioning changes the math

Hand-correction at one to two hours per video, multiplied by a five-thousand-asset catalogue, multiplied by a CAD-50-per-hour staff-or-student-worker rate, produces a six-figure CAD project. Human captioning at CAD 1.50–CAD 4.00 per minute of video, multiplied by an average 30-minute lecture across that catalogue, lands in the same band — sometimes worse.

Glossary-biased captioning is a different cost shape. The institution builds the glossary once. Each minute of video costs a fraction of human-vendor pricing. Accuracy is high enough on the proper-noun surface that the human-review pass collapses to a quick scrub of the amber-highlighted glossary surface. For a 5,000-asset catalogue at an average 30-minute length — 2,500 hours — the GlossCap math (Org plan, retrofit window) lands well under the in-house and vendor-only paths. See the vendor pricing breakdown for the per-hour comparison.

The other cost most retrofit calculations miss is the cost of being non-conformant in the second-cycle ACR filing — the 2029 cycle that follows the 2026 one. Glossary-biased captioning is what stops the proper-noun-mangling recurrence, and is what supports a defensible "fully compliant" attestation in 2026.

FAQ — Brightspace captioning

Does D2L Capture's auto-transcript clear AODA SC 1.2.2 or ADA Title II?

Capture's auto-transcript is in the same 80-90% substantive-accuracy band as every other generic auto-captioner. The AODA IASR's WCAG 2.0 AA bar and ADA Title II's WCAG 2.1 AA bar both enforce captions "that accurately convey the audio." For a no-proper-noun, conversational recording, an auto-transcript can be substantively accurate. For lecture, regulated-content, or technical-procedure recordings, auto-transcripts virtually always need correction. The defensible posture is to treat auto-transcripts as drafts and run a glossary-biased correction pass before the recording is exposed to a student.

What format do I upload to Brightspace's Insert Stuff — SRT or VTT?

Both work in current Brightspace versions. WebVTT is preferred where the captions need positioning cues, styling, or speaker-identification metadata; SRT is the universal default and works on every consumption surface. Most institutional retrofits standardise on SRT for Insert Stuff and let Capture and Panopto/Kaltura output VTT or platform-native formats internally.

How do AODA and ADA Title II differ for a Brightspace tenant operating in both regimes?

The substantive captioning bar is identical — synchronised prerecorded captions that accurately convey the audio. AODA references WCAG 2.0 AA; ADA Title II references WCAG 2.1 AA. The captioning Success Criterion (1.2.2) is unchanged between 2.0 and 2.1. The structural difference is enforcement: AODA pairs WCAG with a three-year compliance reporting cycle (administrative monetary penalties under O. Reg. 432/07 backstop the filings); ADA Title II is enforced through DOJ action and private suit, with no recurring filing. A single caption track can clear both regimes; the documentation differs.

If we use Panopto or Kaltura through the LTI app, whose captioning workflow do we follow?

The hosting platform's. Panopto's captions and Kaltura's captions live in those platforms' libraries and follow the video into Brightspace via the LTI insert. Re-captioning the LTI-embedded portion of the catalogue runs through the Panopto API and the Kaltura API; the asset register on your side has to merge Brightspace and the external platforms.

What does the Accessibility Compliance Authority sampling actually look like?

Per the published Ministry guidance and the Accessibility Compliance Authority's investigative practice: open the institution's public-facing site, find the catalogue, watch a slice of video, read captions against audio, sample several courses to look for a pattern. The 2023 cycle saw heightened scrutiny of post-secondary tenants particularly; the 2026 cycle is expected to be similar with attention to back-catalogue retrofits in tenants that previously attested compliance.

What about Indigenous-language captioning?

AODA does not require Indigenous-language captioning per se. But where a video contains Indigenous speakers or terms — common in Indigenous-studies, cultural-competency, and land-acknowledgement-related content — the captioning still has to "accurately convey the audio." Mangling Anishinaabe, Cree, Inuktitut, or Mohawk terms in a substantive caption is non-conformant under SC 1.2.2 even if the underlying audio is in English. The pragmatic operational answer is to add Indigenous community and language names to the institutional glossary.

Does the EAA reach Brightspace tenants in EU member states?

Yes — D2L has a meaningful European customer base, particularly in higher education and corporate L&D. The EAA's substantive captioning bar references EN 301 549 V3.2.1, which cross-walks WCAG 2.1 AA. See the EAA captions reference. Cross-border tenants run a single glossary-biased caption track to clear AODA + Title II + EAA.