Platform reference · LearnUpon · Customer education LMS · ADA Title III · WCAG 2.1 AA · EAA

LearnUpon captions: customer education LMS, partner training portals, and WCAG 2.1 AA compliance

LearnUpon is a B2B-focused LMS widely used by SaaS companies, professional services firms, and mid-market businesses to deliver customer training academies, partner enablement portals, and internal employee L&D — often all three from a single LearnUpon account using the platform's multi-portal architecture. LearnUpon does not provide built-in auto-captioning for custom video content uploaded to the platform. Organizations must produce and upload an SRT or VTT sidecar file for each video to display captions in LearnUpon's course player. Customer-facing LearnUpon portals accessible to customers, partners, or resellers are public accommodations under ADA Title III, which requires WCAG 2.1 AA-compliant captions on video content. Internal training portals covering employees are subject to ADA Title I employer accommodation obligations. LearnUpon's Irish headquarters and heavy EU customer base place the platform squarely within the European Accessibility Act compliance scope for EU SaaS companies. The primary vocabulary failure surface is company-specific: because customer education video teaches the company's own product, every feature name, product tier, technical term, and integration name is a proper noun that generic speech-to-text systems fail to transcribe accurately — making caption quality a direct reflection of training quality.

TL;DR

LearnUpon does not auto-generate captions for custom video content. Upload an SRT or VTT sidecar file per video in the LearnUpon course builder → Module section → Upload caption file. Customer-facing academies (accessible without login) require WCAG 2.1 AA captions under ADA Title III. Internal training portals require WCAG 2.1 AA captions under ADA Title I for employees with hearing disabilities. EU companies using LearnUpon face EAA obligations by June 2025. The critical accuracy problem: your product names, feature names, and integration vocabulary are exactly the terms generic STT transcribes incorrectly — a domain-specific glossary is required for accurate customer education captions.

LearnUpon's multi-portal architecture and captioning scope

LearnUpon's distinguishing architectural feature is its multi-portal system: a single LearnUpon account can run multiple separate branded portals, each with its own domain, branding, learner audience, and course catalog. A typical SaaS company deployment might include:

Customer portal — publicly accessible or login-required academy for end users of the company's product. This is the primary customer education surface and the highest-priority captioning target for ADA Title III.
Partner portal — controlled-access portal for resellers, value-added resellers (VARs), system integrators, and implementation partners. Partner portals involve a commercial relationship between the company and its channel partners; captioning obligations for hearing-impaired partners apply under the commercial-services analysis even when the portal is not publicly accessible.
Internal portal — employee-facing L&D content for the company's own team. Covered by ADA Title I employer accommodation obligations.

The multi-portal architecture creates a common organizational challenge: captioning decisions must be made separately for each portal's compliance profile. A video asset that exists in both the customer portal (Title III) and the internal portal (Title I) requires captioning for both reasons, but the obligation arises from different legal sources with different enforcement mechanisms.

Portal access configuration and ADA Title III analysis

LearnUpon portals can be configured with different access levels:

Public self-registration — any visitor can create an account and access course content. This is a public accommodation under ADA Title III regardless of whether a fee is charged. Captions are required on all video content.
Invitation-only / admin-assigned enrollment — learners access the portal only after being individually invited. The commercial relationship between the company and its customers or partners still creates an accessibility obligation, but Title III's public-accommodation analysis applies with less certainty. California Unruh Act ($4,000 per incident) and the commercial-services analysis still apply in California regardless of federal Title III analysis.
Free preview content — some LearnUpon portals offer sample lessons or previews accessible without enrollment. If these preview modules include video, the preview video is publicly accessible and Title III applies to the preview content even if the full course requires enrollment.

The practical recommendation: treat all customer-portal video content as Title III-covered and caption it fully, regardless of access configuration. The determination of whether a specific portal instance qualifies as a "public accommodation" is a legal question; captioning all customer-portal video is the operationally simpler and legally safer approach.

How captions work in LearnUpon

Module-level SRT/VTT upload

LearnUpon's course builder organizes content as modules within courses. Video content is added as a video module. To add captions to a LearnUpon video module:

Build or edit the course in the LearnUpon admin interface and navigate to the video module.
In the module editor, locate the caption/subtitle upload option (typically in the module's media settings).
Upload an SRT or VTT caption file corresponding to the video content.
Save the module. The caption track will appear in the LearnUpon course player's caption menu.

LearnUpon does not provide a built-in speech-to-text engine to generate captions automatically. Each video module requires a separately produced caption file. For an organization with hundreds of course videos across multiple portals, the operational scale of producing accurate caption files for all video modules is the primary captioning challenge.

Hosted video vs. embedded external video

LearnUpon supports both hosted video (uploaded directly to LearnUpon) and embedded external video (from YouTube, Vimeo, Wistia, or other video platforms embedded via URL or embed code). Caption management differs by source:

LearnUpon-hosted video — captions are managed via the SRT/VTT upload in the LearnUpon module editor as described above.
YouTube-embedded video — captions are managed on the YouTube video settings page. YouTube's auto-generated captions fail on company-specific vocabulary and must be corrected or replaced. See the LearnUpon module settings to ensure the embedded YouTube player displays the caption track.
Vimeo-embedded video — captions are managed in Vimeo's video settings → Accessibility. SRT or VTT files uploaded to Vimeo display in the embedded player. See Vimeo captions for training videos for the Vimeo caption workflow.
Wistia-embedded video — captions are managed in Wistia's video settings. See Wistia captions for the Wistia caption workflow. Wistia is common for customer education video at SaaS companies because of its analytics and gating features; Wistia-embedded LearnUpon modules require caption management at the Wistia level.

SCORM and xAPI packages

Organizations using authoring tools like Articulate Storyline, Articulate Rise, Adobe Captivate, or Lectora to build SCORM or xAPI packages for LearnUpon courses should manage captions at the authoring tool level. SCORM packages that include video with captions built into the package display those captions when the SCORM module runs inside LearnUpon's player. The alternative — relying on a post-authoring caption layer at the LMS level — is generally less reliable for SCORM-packaged content because the video is embedded inside the package.

ADA compliance for LearnUpon customer and partner portals

ADA Title III: customer-facing academy video

ADA Title III (42 U.S.C. § 12182) prohibits discrimination on the basis of disability in places of public accommodation. Courts have consistently held that commercial websites and online services qualify as places of public accommodation when they operate as a "nexus" between the business and its customers — exactly the function LearnUpon customer academies serve. A customer education portal that teaches users how to use a software product, complete a certification, or onboard successfully is providing a service to those customers; denying access to that service to a customer with a hearing disability through absent or inaccurate captions is a Title III violation.

The damages exposure that most motivates action: California Unruh Civil Rights Act, which provides for $4,000 in statutory damages per violation (per incident of encountering inaccessible content) for California-based users. For a SaaS company with California customers — which is nearly every SaaS company — each uncaptioned customer academy video accessible to California users creates a potential $4,000 per-incident damages exposure.

ADA Title I: internal employee portal

ADA Title I (42 U.S.C. § 12112) requires employers with 15 or more employees to provide reasonable accommodations for employees with disabilities, including providing accessible training materials. LearnUpon internal portals containing mandatory training, role-based onboarding, or required compliance training must provide accessible video with accurate captions for hearing-impaired employees. The accommodation obligation is strongest for required training — training that employees must complete as a condition of employment or that determines certification, advancement, or compensation — and extends to all employees, including remote employees in all states.

Section 1557 for healthcare SaaS companies

SaaS companies that sell products or services to healthcare providers and that receive federal financial assistance (for example, through government contracts, grants, or participation in federal healthcare programs) may be covered entities under ACA Section 1557 (42 U.S.C. § 18116). The 2024 HHS final rule (89 Fed. Reg. 37,522) requires covered entities to provide effective communication to individuals with disabilities, including through accessible video, under 45 CFR § 92.207. For healthcare SaaS companies using LearnUpon for customer academies that train healthcare provider customers, Section 1557's effective-communication standard applies alongside Title III. See Section 1557 captions for the detailed compliance analysis.

EAA and international compliance for LearnUpon

European Accessibility Act (EAA)

The European Accessibility Act (Directive 2019/882) has been enforceable since June 28, 2025. LearnUpon is headquartered in Dublin, Ireland, and serves a large EU customer base — making EAA particularly relevant for both LearnUpon as a platform vendor and for LearnUpon customers operating in the EU. The EAA's accessibility requirements for digital services under Article 4 mandate that video content include synchronized captions meeting the EN 301 549 V3.2.1 standard (Clauses 7.1.1–7.1.5). For EU SaaS companies using LearnUpon for customer academies serving EU customers, the EAA is the primary compliance driver with national transpositions in all 27 EU member states enforceable as of June 2025. See EAA captions requirements for the full EAA analysis.

AODA for Canadian LearnUpon deployments

Canadian organizations with 50 or more employees are covered by Ontario's Accessibility for Ontarians with Disabilities Act (AODA) Integrated Accessibility Standards Regulation O. Reg. 191/11 § 14, which requires new and updated training video to include captions meeting WCAG 2.0 AA. LearnUpon is used by many Canadian SaaS and professional-services companies. For AODA-covered organizations, new customer academy or internal training video uploaded to LearnUpon must include captions, and back-catalogue content must be captioned according to the remediation obligations. See AODA captions for the Ontario compliance timeline.

The company-vocabulary failure mode in customer education video

Customer education video has a structural vocabulary challenge that makes it uniquely sensitive to STT accuracy problems: the content being taught is the company's product. Every feature name, interface element, API endpoint name, integration name, product tier, and configuration term is a proper noun. These are precisely the terms that are absent from general speech-to-text language model training data — because your company's product vocabulary does not appear in Wikipedia, Common Crawl, or any general text corpus in significant volume.

Product vocabulary failure patterns in customer education

Common STT failure patterns in customer education video include:

Feature names with internal capitalisation — "SmartSync," "QuickCapture," "AutoFlow" — are typically split by STT into "smart sync," "quick capture," "auto flow" and lowercased. In a tutorial video explaining "click the SmartSync button to synchronize your data," the caption reading "click the smart sync button" loses the product vocabulary precision that the tutorial is designed to establish.
API and technical terms — REST endpoint paths, SDK method names, and configuration parameter names are typed text that speakers read aloud in tutorial videos. "Call the /v2/users/sync endpoint" may be transcribed as "call the slash V 2 slash users slash sync endpoint" or garbled in a variety of ways depending on how the speaker articulates path separators and version numbers.
Integration partner names — Customer education video for a SaaS product commonly includes integration tutorial modules covering connections to Salesforce, HubSpot, Slack, Jira, Zapier, and other integration partners. Partner company names that are common words or unusual spellings fail systematically: "Zapier" → "zap-ier" (incorrect pronunciation rendering), "Slack" → usually correct, "HubSpot" → occasionally "hub spot" (split). The integration partner's own product vocabulary then adds a second vocabulary layer on top of your product vocabulary.
Pricing tier and plan names — "Starter," "Professional," "Enterprise," "Business," "Growth" as product plan names are common English words, but "TeamPlan," "GrowthPlus," "EnterpriseMax" as invented product tier names fail at the same rate as other product names.
Customer-specific vocabulary — LearnUpon supports white-labeled portals and custom course catalogs for customers. Some organizations teach their customers workflows that incorporate the customer's own internal terminology (their data model names, their configuration templates). This creates a third vocabulary layer — the customer's vocabulary inside the vendor's vocabulary — that is entirely outside any generic STT training data.

Why accuracy matters more in customer education than in other training

In internal compliance or soft-skills training, a caption error produces an inconvenience: the learner gets a slightly wrong word in a low-stakes context. In customer education video, a caption error produces a product comprehension failure: the learner reads an incorrect feature name and either (a) cannot follow the tutorial step, (b) searches for a UI element that does not exist under the name shown in captions, or (c) incorporates the wrong term into their mental model of the product and uses the wrong term when they contact support or describe the feature to colleagues. For a hearing-impaired customer who relies on captions rather than audio to follow customer education video, a systematic pattern of STT failures on product vocabulary makes the training effectively inaccessible — not because captions are absent, but because the captions consistently fail on the most important words in the content.

Caption workflow for LearnUpon customer academies

Building a product glossary for LearnUpon customer education

Before generating captions for any LearnUpon customer education video, build a glossary that covers:

All product names, feature names, and UI element names in your product (especially any with non-standard capitalisation or portmanteau constructions)
All pricing tier names
All integration partner names and their primary product names
All API endpoint paths, SDK method names, and configuration parameters used in tutorial content
Technical terms specific to your product category (the industry vocabulary your product operates in)
Any certification or credential names your LearnUpon academy awards

This glossary, loaded as a strong prior in GlossCap's glossary-biased Whisper transcription, produces captions that use your product vocabulary accurately — resolving the STT failures that make generic auto-captions unreliable for customer education content.

Prioritizing the LearnUpon back-catalogue

For organizations with existing LearnUpon portals and a large library of uncaptioned video, a triage framework helps prioritize captioning effort:

Required / mandatory courses first — courses that customers or employees must complete for certification, continued access, or compliance are the highest compliance risk.
High-enrollment courses second — LearnUpon's reporting shows course enrollment and completion rates. High-enrollment courses affect the most learners and carry the greatest aggregate risk exposure.
Publicly accessible portal content third — content in LearnUpon portals with public self-registration is the most legally exposed under Title III.
New content on an ongoing basis — establish a workflow for captioning all new video modules before they are published to the portal, so the back-catalogue is the only outstanding remediation work.

See GlossCap pricing

FAQ — LearnUpon captions

Does LearnUpon auto-generate captions for uploaded video?

LearnUpon does not provide a built-in speech-to-text auto-captioning engine for custom video content uploaded to the platform. Organizations must produce SRT or VTT caption files using an external captioning service or transcription workflow and upload those files to each LearnUpon video module. Verify the current capabilities in your LearnUpon portal's Help documentation, as platform features evolve — but as of mid-2026, the standard workflow for customer-uploaded video content is manual caption file upload rather than auto-generation. For video embedded from external platforms (YouTube, Vimeo, Wistia), captions are managed at the source platform and display in the LearnUpon embedded player based on the source platform's caption track.

Our LearnUpon academy is invite-only — do we still need captions under Title III?

The Title III public-accommodation analysis for invite-only portals is less clear-cut than for publicly accessible portals, but the practical answer for most SaaS companies is: yes, you still need captions. First, the commercial relationship between a SaaS company and its paying customers creates an accessibility obligation even for invitation-only portals — denying a hearing-impaired customer full access to the product training they paid for (either directly as part of the subscription, or indirectly as part of the product's value proposition) raises both Title III and general commercial-service accessibility arguments. Second, California Unruh Act ($4,000 per incident) does not require the portal to be "public" in the Title III sense — it applies to any business providing services in California, which includes invite-only customer portals serving California customers. Third, EAA for EU customers and AODA for Canadian customers do not turn on the public-vs-private distinction in the same way Title III does. The practical recommendation is to caption all customer portal video regardless of access model.

We have 300 course videos in LearnUpon — how do we prioritize captioning?

Use LearnUpon's reporting to export course enrollment data and completion rates. Sort by enrollment volume descending. The highest-enrollment courses affect the most learners and carry the greatest aggregate legal exposure — certificate or required-completion courses first, then the most actively used customer onboarding courses, then the broader library. Simultaneously, establish a new-content workflow so that any video added going forward is captioned before publication — this stops the back-catalogue from growing. The back-catalogue remediation then becomes a finite project (even if a large one) rather than an ongoing expanding gap. For the back-catalogue, GlossCap's batch processing with a shared product glossary produces consistent vocabulary results across all videos in the same product area, since the product vocabulary is common across all customer education modules for the same product.

What file format does LearnUpon accept for captions?

LearnUpon accepts SRT (.srt) and WebVTT (.vtt) caption files for video modules. Both formats are text-based time-coded subtitle files. SRT is the simpler format (sequential index, start/end timestamps, caption text); WebVTT adds metadata capabilities, speaker tags, and positioning cues that are not used in most learning management contexts but may be useful for multi-speaker content. For LearnUpon's standard course player, either format works; SRT is slightly more universally supported across third-party players and export workflows, making it the safer default if you are unsure which format to produce. GlossCap exports both SRT and VTT from each transcription job — choose the format that your LearnUpon module editor requires or that your content workflow prefers.

Our LearnUpon academy is in multiple languages — how does captioning work?

LearnUpon supports multi-language course content, and each language version of a video module requires its own caption file in the corresponding language. If your customer education videos are produced in English and localized to French, German, Spanish, or other languages, the localized video versions require localized caption files — not just translated text, but time-coded caption files that match the timing and phrasing of the localized narration. GlossCap's glossary-biased Whisper approach is language-aware: Whisper natively supports transcription in 99 languages, and the glossary entries for your product vocabulary can be configured per language (including mixed-language glossary entries for product names that should remain in English across all localizations). The most common localization captioning workflow: produce the English caption file first, then use a qualified translator to produce localized SRT/VTT files that align with the localized audio timing.