LMS integration
Absorb LMS captions: the SRT upload flow and what Absorb's accessibility posture requires
Absorb LMS is the enterprise-mid-market LMS most common in regulated industries — healthcare systems, financial services, government contractors — exactly the segments most affected by the ADA Title II deadline that went live 2026-04-24. Absorb has published an accessibility posture that explicitly references WCAG 2.1 AA; this is what caption upload looks like inside an Absorb lesson, how to retrofit a large library without breaking learner-progress state, and why GlossCap is the right upstream for regulated training content.
TL;DR
In Absorb LMS, captions attach per video lesson (sometimes called a "video object" depending on your console version). The upload control is on the lesson-edit page, typically under a "subtitles", "captions", or "accessibility" section. Absorb accepts SRT broadly; VTT works on newer player builds. Absorb's own product-level accessibility docs reference WCAG 2.1 AA — which makes the caption content quality, not the platform, the load-bearing piece. Glossary-biased captions from GlossCap preserve the product names, drug names, and SDK symbols that auditors actually sample.
Where the caption upload actually lives in Absorb
Absorb's admin console has evolved across major versions; the high-level model has been stable: a course contains chapters, chapters contain lessons, and a video lesson has a video file plus optional accessibility metadata. The caption upload is on the lesson edit page.
Look for:
- A "Captions" or "Subtitles" section on the lesson editor, typically under or near the video upload control.
- An "Accessibility" tab or collapsible panel that bundles caption upload with transcript upload and alt-text fields.
- A language dropdown — Absorb supports multi-language caption tracks on a single video lesson.
If the control is not immediately visible, check whether your console is on an older theme where the caption-upload is tucked inside an "Advanced" expander. Absorb has been incrementally moving caption controls up in the UI as accessibility compliance has become mandatory across their customer base.
Absorb's accessibility posture and what it means for you
Absorb has published product-level accessibility commitments referencing WCAG 2.1 AA. In practice this means:
- The Absorb video player exposes a standard captions/CC control.
- Keyboard navigation, screen-reader compatibility, and colour-contrast ratios on the player chrome are targeted at WCAG 2.1 AA by the vendor.
- But: the player's compliance posture has zero effect on whether your content passes an audit. If the video has no caption file attached, or the attached captions mangle your terminology, the audit finding is against the content, not the LMS.
That split — "Absorb's player is accessible; your content may not be" — is the single thing training-ops leads underestimate when picking a compliance posture. The SC 1.2.2 page walks through what an auditor actually samples on the content side.
The retrofit workflow for an Absorb library
For a 100–500-video Absorb library (typical mid-market and regulated-industry footprint), the right retrofit order is:
- Generate the lesson inventory. Absorb's reporting surface lets admins export lessons filtered by type = video. Pull course / chapter / lesson / video-filename / duration into a spreadsheet.
- Classify by compliance scope. Learner-facing compliance training (HIPAA, SOX, GDPR, anti-harassment), customer-facing product training, and ADA-covered onboarding all need captions first. Internal admin walkthroughs can follow. Prioritise by deadline pressure and audit risk.
- Caption in GlossCap with one glossary. Upload source videos to a GlossCap batch, sync your company glossary once (Notion / Confluence / Google Docs or pasted list). Every module inherits the same terminology model; the SDK symbols and drug names and product names stay consistent across the library.
- Download SRTs with lesson-matching filenames. Configure the batch export to use a filename pattern tied to your Absorb lesson IDs so the upload step is drag-and-drop.
- Upload per lesson in Absorb. Open each video lesson, attach the matching SRT. For a large library this is a focused 1–3-day admin sprint depending on tier.
- Verify as a non-admin learner. Log in as a test learner on each customer-facing domain / portal variant your Absorb tenant runs. Open 5 sampled lessons, confirm captions render, verify the CC toggle works. This catches lesson-upload-to-wrong-asset errors.
The learner-progress-state gotcha
Absorb tracks learner progress (SCORM- / xAPI-compatible) at the lesson level. A video lesson that a learner has partially watched has a progress record tied to it. Admins occasionally assume that attaching a caption file to a lesson "updates" the lesson in a way that might invalidate learner state. It does not. Caption attachment is metadata; it does not replace the lesson's video asset, so progress state is unaffected.
The move that does break learner state is deleting and re-creating the lesson — which is what some admins do when they can't find the caption control. If you can't find caption upload on an existing lesson, ask your Absorb customer success contact for the admin console path before doing anything destructive.
Why glossary-aware captioning matters more on Absorb
Absorb's customer base is disproportionately regulated industries: healthcare, financial services, government, life sciences. The training content in these verticals is terminology-dense in ways that general-purpose speech models handle badly. Drug names (tirzepatide, semaglutide, empagliflozin), procedure names, regulatory acronyms (HIPAA, GDPR, FINRA, FedRAMP), internal product names, and SDK symbols are exactly what YouTube auto-caption, Otter, and Whisper-base all garble.
Auditor sampling on healthcare compliance training is not random — it is targeted at the segments where drug names are spoken, because that is where clinical mis-captioning has real downstream cost. Same on financial compliance: the auditor samples segments where regulatory acronyms appear. If those specific words are wrong on sampled segments, the finding is against your training, not the LMS.
GlossCap's one differentiator fixes exactly this case: your company glossary and vertical-specific term list get logit-boosted into the Whisper-large decoder before any output is generated. When the acoustic model is ambiguous between "tier zip a tide" and "tirzepatide", the decoder has already been nudged toward the right surface form. The output SRT is WCAG-compliant on first export.
Related questions
Does Absorb support VTT?
Recent Absorb player builds accept VTT alongside SRT. If you're on an older tenant, SRT is the safe default — see our SRT page. If you need VTT's styling hooks, contact your Absorb admin to confirm the player build supports it.
Can I bulk-attach captions via the Absorb API?
Absorb exposes a REST API (commonly called AbsorbOne or the Absorb API) covering lesson and course CRUD. Caption attachment on video lessons is supported by the API in recent versions; check your tenant's API docs for the exact endpoint shape. For one-time retrofits, the admin UI is usually faster. For ongoing ingestion, API-driven attachment is the right path.
What about SCORM packages with embedded video?
SCORM packages bundle content separately — the caption file is inside the package ZIP and travels with the SCORM upload. Retrofitting a SCORM package means re-authoring the package with the caption file included and re-uploading; Absorb itself does not let you retrofit captions on a SCORM-embedded video from the lesson editor. For non-SCORM video lessons, the lesson-level caption upload works.
Does Absorb meet WCAG 2.1 AA on its own?
Absorb publishes accessibility commitments targeting WCAG 2.1 AA on the player and admin chrome. Your content is your responsibility — and that is where glossary-aware captioning is the load-bearing piece. Our WCAG 2.1 AA reference breaks the compliance split.