Tool reference · Articulate Storyline 360

Articulate Storyline captions: glossary-biased VTT/SRT for the dominant L&D authoring tool

Articulate Storyline 360 is, by an uncomfortable margin, the dominant authoring tool in corporate L&D. The Articulate 360 suite — Storyline 360 plus Rise 360 plus the Articulate Review and content libraries — is what the vast majority of 50-to-500-employee training-operations teams build slide-based interactive training in. Storyline produces a SCORM, xAPI, AICC or HTML5 package that lands in TalentLMS, Docebo, Absorb, Healthstream, Cornerstone OnDemand, or any other compliant LMS as a single course asset, with caption tracks embedded in the package. The captioning question on Storyline is not whether the tool supports captions — it does, and well — it is whether the caption text inside the published package preserves the vocabulary the training was meant to teach. Generic auto-captioning fails this surface predictably, because Storyline modules carry the exact density of product names, SDK symbols, drug names, and policy codes that ASR has the least training data for.

TL;DR

Storyline 360 supports closed captions on every slide-level audio and video object via WebVTT (preferred) or SRT import, plus its in-tool slide-level caption editor for direct authoring. The published output (SCORM 1.2, SCORM 2004, xAPI, AICC, HTML5, or web object) carries the captions inside the package — an LMS administrator does not separately upload caption files; the LMS plays the package as-is. The technical caption-import is straightforward; the difficulty is upstream — the imported VTT file. Hospital, SaaS, engineering, and university training built in Storyline carries high proper-noun density (drug names, SDK terms, regulatory acronyms, product names, internal codes), and generic STT mangles these systematically. Glossary-biased captioning with the customer's controlled vocabulary as the project glossary preserves these surface forms on first export, so the imported VTT lands clean and the published package holds up at audit.

What Storyline 360 is, and where captioning sits in the publish path

Storyline 360 is Articulate's slide-based interactive authoring tool, distinct from Rise 360 (the responsive microlearning tool — see our Articulate Rise captions reference). Storyline produces interactive courses with branching, variables, triggers, software simulations, and quizzing, and is positioned for training that needs higher production value than Rise's block-based approach allows.

The publish artefact is a single zip (SCORM 1.2, SCORM 2004, xAPI / Tin Can, AICC, HTML5-only web, or Articulate Review) containing:

An index_lms.html entry point and the SCORM/xAPI runtime shim that talks to the LMS
The slide content as a sequence of HTML/JS/CSS plus media assets (audio, video, images)
Caption files in WebVTT format alongside each slide-level media asset, with the per-slide closed-caption toggle wired to the LMS player
The course manifest (imsmanifest.xml for SCORM, tincan.xml for xAPI) declaring the assets and the entry point

The captions are inside the package. There is no separate "upload a caption file to the LMS" step; the LMS plays the published package, the published package plays the captions. This is the architectural reason Storyline captions land in the upstream authoring stage, not the downstream LMS administration stage.

The Storyline caption-import mechanic in detail

Storyline 360 supports caption import in two formats and three workflows:

WebVTT import (preferred). The slide-level audio or video object's properties pane has an "Import captions" entry that accepts .vtt files. WebVTT is the recommended format because it preserves cue identifiers, region/positioning hints, and styling cues that survive into the published HTML5 output. The import maps cues to slide-time directly.
SRT import. The same import path accepts .srt. SRT is simpler — no styling, no regions — but reliably converts to WebVTT under the hood. For monolingual training without complex positioning needs, SRT is fine.
In-tool caption editor. Storyline ships its own slide-level caption editor with a waveform view of the audio/video and a row-per-cue editing pane. This is the path for last-mile correction after import — fix the proper-noun mangle, retime a cue, split a long cue.

The publish dialog has the "Closed captions" player feature toggled on by default in the Modern Player. The course player exposes a CC toggle button, and the LMS plays the captioned course directly without any LMS-side configuration.

The proper-noun failure mode on Storyline-authored training

Storyline modules concentrate vocabulary density because of how they are typically built. The slide-level audio is recorded by a SME or narrator reading a script that was written from source material — engineering documentation, policy-and-procedure libraries, drug-formulary references, regulatory text. The script preserves the surface forms; the recording preserves the surface forms; auto-captioning the recording is the step that breaks them.

Predictable failure surfaces by training type:

Engineering onboarding. SDK names, library names, command-line tools — kubectl, pytest, terraform, tsc, esbuild, cargo, helm, argocd. See our engineering onboarding captions reference for the full failure-mode catalogue.
Medical / clinical training. Drug generic and brand names, procedure terms, regulatory acronyms — tirzepatide, semaglutide, dexmedetomidine, EMTALA, NPSG.03.05.01. Our medical training captions reference walks the drug-name failure mode in depth.
Compliance training. Regulatory acronyms, framework names, citations — SOX § 404, HIPAA § 164.530(b), GDPR Art. 32, ISO 27001 Annex A. The compliance training captions reference covers the framework crossover.
Software / product training. Product names, feature names, UI element names — the SaaS-vendor catalogue. Customer-academy training built in Storyline mangles these; sales-enablement training built in Storyline mangles these. See our sales enablement captions reference.
Safety training. Chemical names from the SDS, OSHA citations, MSHA standards — safety training captions reference walks the OSHA / MSHA / HazCom surface.
Locally meaningful organisational vocabulary. Internal product codes, programme names, team names, system names. Generic STT has zero exposure to these; the customer's glossary is the only source.

The shared pattern is that the words a Storyline module is built to teach are exactly the words generic ASR has the least exposure to. Glossary-biased captioning closes that gap upstream of the Storyline import.

The glossary-biased workflow for Storyline-authored training

Pull the customer's controlled vocabulary. Engineering teams: the SDK reference, internal architecture wiki, command-line glossary. Healthcare: drug formulary, policy-and-procedure index, EHR-specific terminology. SaaS: product-feature catalogue, customer-academy term list, competitor-name register. Compliance: framework register, citation index, internal control names. Most organisations have a glossary in some form already; loading it into the captioning workspace is a one-time setup.
Process the slide-level audio export. Storyline exports per-slide audio as standalone WAV or MP3. Caption that audio with the workspace glossary biasing the decoder, producing one VTT file per slide. The glossary preserves the proper-noun surface forms on first export.
Reviewer pass with amber-highlight UI. Every glossary-applied term is highlighted with source-line provenance — drug-formulary version, SDK reference URL, policy-document location. The reviewer scrubs each slide's captions in minutes, with corrections feeding the workspace glossary for the next module.
Import the VTT files into Storyline. Slide by slide, via the audio/video properties pane. The in-tool caption editor handles last-mile retiming and cue-splitting. The publish artefact carries the captions inside the package.
Publish to the target LMS. SCORM 1.2 for legacy LMS targets, SCORM 2004 / xAPI for modern targets, HTML5-only for direct embed. The published package carries the captions; the LMS plays them without separate configuration.
Document the captioning provenance. Each published course gets a row in the asset register: caption source (vendor + glossary version), reviewer name and role, review date, glossary term count. ADA Title II and Section 508 audits read this register as evidence.

See pricing

How Storyline-published captions land in the major LMS targets

The published Storyline package is portable across LMS targets. The captions ride along regardless of where the package lands. The LMS-specific notes:

TalentLMS — SCORM 1.2 / SCORM 2004 / xAPI accepted; the player respects the embedded CC toggle. SMB/mid-market default for Storyline-authored content.
Docebo — full SCORM and xAPI; the Docebo player reads the embedded captions with no additional configuration. Mid-market and enterprise default.
Absorb LMS — SCORM 1.2 / SCORM 2004 / xAPI / AICC. Absorb's Engage player respects the published Storyline CC toggle.
Kaltura — when Storyline content is wrapped into a Kaltura course-launch object, captions ride inside the SCORM package; for direct video assets, separate caption upload via Kaltura's REST API is the path.
Healthstream — for hospital-authored Storyline content, the SCORM package carries captions into the hStream content authoring tools without further configuration.
Panopto — for academic medical centres and universities mixing lecture capture with Storyline-authored compliance modules; the SCORM package and Panopto video coexist with their own captioning paths.

The portability of the embedded-caption pattern is exactly the reason the captioning workflow lands in the upstream Storyline authoring step, not in the downstream LMS configuration.

Storyline's "Generate captions automatically" feature — and why it's not the answer

Storyline 360 ships a built-in caption-generation feature that runs ASR over the slide-level audio and produces a draft caption track in-place. It is genuinely convenient for narrator-recorded slide audio. It is also exactly the failure surface this page is about.

The built-in ASR does not have access to the customer's drug formulary, SDK reference, or product-feature catalogue. It mangles the proper nouns the training was built to teach. The published package carries the mangled captions into the LMS, and the L&D team discovers the failure mode either from a learner complaint or from an audit finding. The hand-correction cost — the half-FTE captured in the L&D budget that nobody named — is real, and our long-form post on the hidden half-FTE walks the math.

The glossary-biased path solves the upstream problem: the generated VTT lands clean the first time, the reviewer pass is minutes-per-module rather than hours-per-module, and the published package's captions hold up at audit.

The Storyline back-catalogue retrofit pattern

The back-catalogue question for Storyline is operationally distinct from the back-catalogue question for an LMS-hosted video file, because the Storyline source .story file is the editable source of truth — re-publishing a corrected caption track requires opening the source, importing the new VTTs, re-publishing, and re-uploading the SCORM/xAPI package. Operational notes:

Source-file recovery. The first prerequisite for the retrofit is having the .story source files. Many organisations lose track of these over staff turnover; the search-and-recover step is sometimes the longest part of the project.
Bulk-export slide audio. Storyline supports per-slide audio export through Articulate Review or the source file's media library. For large courses (100+ slides), a scripted export to a folder structure that mirrors the slide hierarchy is the typical first step.
Glossary-biased re-captioning. The exported audio runs through the captioning workflow with the customer glossary biasing the decoder. Output: one VTT per slide.
Bulk-import into the source file. Storyline's caption-import is per-slide; for large courses, the bulk-import via the in-tool batch importer or a file-system-level VTT placement (Articulate's storage layout permits this for advanced workflows) is the practical path.
Re-publish and re-upload. Each course re-publishes to the same SCORM/xAPI target and replaces the existing course in the LMS. Most LMSes preserve learner progress on re-upload of an updated SCORM package; verify this on your LMS before bulk-replacing.

The end-state is a Storyline back-catalogue with audit-clean captions, a captioning provenance log per course, and a glossary that has compounded across the catalogue.

How Storyline captions intersect ADA Title II and Section 508

Storyline-authored training in scope of ADA Title II (state and local government, including public universities and county hospitals) must meet WCAG 2.1 AA from 2026-04-24. Storyline-authored training in scope of Section 508 (federal government and federal contractors with access requirements flowed down) must meet the EN 301 549 / WCAG 2.0 AA technical bar referenced by 36 CFR § 1194.

The technical caption requirement at WCAG SC 1.2.2 (Captions, Prerecorded) is satisfied by a Storyline package whose embedded captions accurately convey the spoken content. The "accurately" part is the audit-relevant question. For both regimes, the audit-relevant evidence is the captioning provenance log: who captioned, what glossary version, who reviewed, when. The caption files themselves carry the answer; the provenance log is the documentation a reasonable auditor reads.

For private-sector employers, the indirect ADA Title III posture (driven by case law, currently on a 2.0 / 2.1 AA technical reading) plus the European Accessibility Act for EU-operating organisations apply the same WCAG technical bar. See our EAA captions requirements reference for the EU regime and our EAA Q3 2026 inflection-point post for the enforcement landscape.