Lifetime Deal: $199 one-time — PRO forever41d 12h 11m 10s87/100claimed·13 leftClaim your spot →
For instructional designers and L&D teams . May 2026
Course voiceover without a recording booth
Five stages from script to LMS, mapped to the right TTS tool at each step. Plus how the popular authoring tools (Storyline, iSpring, Captivate) play with AI voice, and the volume math for indie creators vs corporate L&D teams. No comparison tables. Just the workflow.
Last updated by the FreeTTS team · Sources at the bottom
Global eLearning market (2026)
$352.98 billion, growing at 19.9% CAGR — AI voice narration saves 60-90% vs hiring human voice actors
Cheapest commercial-ready stack
FreeTTS PRO at $19/month — 1M chars (~14 hours audio), 2,250+ voices, full commercial license, SSML support
SCORM compatibility
All major authoring tools (Storyline, iSpring, Captivate, Rise 360) accept MP3; bundles into the SCORM .zip
Updating course audio
Change one slide script, regenerate that clip, swap the file — done in 60 seconds vs $200+ for human re-recording
Best for talking-head video courses
Synthesia ($30/mo) generates AI presenter with synced voice — no filming needed
Multilingual course narration
FreeTTS PRO covers 149 languages; ElevenLabs Creator supports 32 languages with the same voice across all
The workflow
Script to LMS in five stages
Each stage takes a different tool. Knowing the boundaries saves hours.
1
Script
Write per-slide or per-screen scripts in plain text. Keep sentences short. Spell out acronyms (SCORM, xAPI). Mark intentional pauses with SSML <break/> tags.
Tools: Notion / Google Docs / VS Code
2
Generate audio
Run scripts through your TTS tool. Use one voice consistently across the course so learners settle into it. Test the first 30 seconds before generating the whole batch.
Tools: FreeTTS PRO / ElevenLabs / Murf
3
Edit + clean
Trim silence at start/end. Add room tone if voice sounds too dry. Adobe Podcast Enhance Speech (free) adds warmth that masks the slightly clinical TTS sound.
Tools: Audacity / Adobe Podcast
4
Sync to slides
Drop audio into your authoring tool. Set per-slide playback. For Articulate Storyline use the Audio Track. For iSpring set on the slide narration tab.
Tools: Storyline / iSpring / Captivate
5
Publish to LMS
Export as SCORM 1.2 or 2004 (most LMSs) or xAPI (newer LMSs). Upload the package. Test on at least one learner device before launch.
Tools: Moodle / Canvas / Cornerstone / Workday
Authoring + LMS
How AI voice plays with the popular authoring tools
Each card: how to add audio, how it exports, and what learners get on the LMS side.
Articulate Storyline
SCORM 2004
Drop the MP3 into Audio Track per slide, set timing, publish as SCORM. Audio bundles into the .zip package. Works with every modern LMS.
PowerPoint-based authoring. Add per-slide narration via the Narration Editor, sync to slide timing, publish as SCORM or video. Easiest learning curve if your team already uses PowerPoint.
Strongest for branching scenarios and software simulations. Audio attached per slide, branching object, or background. Steeper learning curve than Storyline or iSpring.
Different category. AI-generated presenter with synchronized voice. Best for talking-head explainers without filming. Pricier ($30/mo). Voice is included so no separate TTS needed.
Articulate's lightweight web-only authoring. Add audio per block. Works well with FreeTTS-generated MP3 narration. Publishes directly to Articulate Review or LMS.
Skip the authoring tool. Upload your TTS audio MP3 directly to a Moodle Lesson or Canvas Module as a media file. Cheapest path for budget-constrained instructors.
Add audioInsert media on pageExportN/A (lives in LMS)Best fitUniversities + small teams
The TTS picks
Which TTS to use at stage 2
Different tools for different course volumes and budgets.
FreeTTS PRO
$19/mo
Best value for indie course creators and small L&D teams. 1M chars/month covers ~14 hours of finished audio. 2,250+ voices in 149 languages, full commercial license, SSML support for technical pronunciation.
1M charsVolume2,250+Voices75+Languages
ElevenLabs Creator
$22/mo
Best voice realism. Voice cloning included. Strong fit when you need your own voice (CEO message, founder course intro) or character voices for compliance training scenarios.
100k charsVolume5,000+Voices32Languages
Murf
$23/mo
Voiceover + slide editor in one tool. Saves a separate authoring step if you make slide-based explainer courses. Good fit for marketing teams producing internal training.
~150k charsVolume120+Voices20+Languages
FreeTTS Creator
$39/mo
Mid-size L&D teams. 5M chars/month covers ~70 hours of audio (roughly 30 short courses or 10 long ones). Same 2,250+ voices, MP3-HD export, batch processing for course-set generation.
5M charsVolume2,250+VoicesHDMP3 export
The math
Volume math by team size
Stop guessing at the right plan. Here's what each tier actually fits.
Solo / Hobby
FreeTTS Free
$0
Personal use only, audio tag at clip start. Fine for testing voices and prototyping a course. Move to PRO before publishing for sale.
Indie creator
FreeTTS PRO
$19/mo
Up to ~14 hours of audio per month. Covers 4-6 short courses or 1-2 long ones with retakes. Most popular plan for Udemy/Teachable creators.
L&D team (5-20 people)
FreeTTS Creator
$39/mo
Up to ~70 hours of audio per month. Covers a course-a-week pace with batch processing for compliance refresh cycles.
Enterprise / Multi-team
Multiple Creator seats
$200+/mo
Mid-size L&D shops typically run 5-10 Creator seats with shared voice consistency. Reach out at /contact for volume pricing.
FAQ
Questions instructional designers ask
Yes. Generate the audio as MP3, drop it into Articulate Storyline / iSpring / Adobe Captivate / Lectora, then publish as SCORM 1.2 or 2004. The audio is bundled into the package and plays through any SCORM-compliant LMS (Moodle, Canvas, Cornerstone, Workday, Talent LMS, dozens more).
Most won't, on neural voices. Microsoft Azure neural voices (powering FreeTTS) and ElevenLabs flagship voices are mostly indistinguishable from human in casual listening. Where learners notice: long-form (45+ min modules), high-emotion content, and complex pronunciation of technical terms (where SSML phonetic spelling helps).
FreeTTS PRO at $19/month for 1M characters (~14 hours of audio) is the cheapest commercial-ready option for high-volume creators. ElevenLabs Starter at $5/month covers 30k characters which is enough for 2-3 short modules.
Yes, with a commercial-license TTS plan. Udemy specifically allows AI voiceover. Teachable, Thinkific, Kajabi, and Skool also allow it. Coursera and edX have stricter content-quality reviews; AI voice is allowed but the course content itself needs to be original and substantial.
Pick one voice, lock the speed and pitch, generate all audio in one batch. Don't mix voices mid-course (jarring for learners). FreeTTS PRO and ElevenLabs both let you pin a voice as the default for an account.
Yes, that's actually the killer use case for AI voice over human narration. Need to update one slide because a stat changed? Edit the script, regenerate that one clip, swap the file. Done in 60 seconds. With a human narrator, this is a $200 fix.
Use SSML phoneme tags or write the acronym phonetically. For example, SCORM becomes 'skorm' in the script if the engine reads it as letters. ElevenLabs has a pronunciation editor. FreeTTS supports SSML phoneme tags for surgical fixes.
Synthesia ($30/mo) generates an AI presenter with synchronized voice. Heygen ($29/mo) is similar. Use these when you need an on-screen instructor and don't want to film. For voice-only narration over slides, FreeTTS PRO is far cheaper.
Yes. FreeTTS PRO covers 149 languages. ElevenLabs supports 32 languages with the same voice across all of them (powerful for global L&D teams). For corporate compliance training in 5 languages, this saves weeks of localization.
Generally no. WCAG and Section 508 don't differentiate between human and AI voiceover. The requirement is captions/transcripts (which you should have regardless) and reasonable audio quality. AI voice meets both.
TTS: 90 minutes total (script generation + cleanup + sync). Voice actor: 5-10 days of back-and-forth. Cost: $19 vs $500-1,200. The trade-off is the personality of a great human narrator vs the speed and updateability of TTS.
PRO at $19/mo handles most indie and small-team needs. Creator at $39/mo (5M chars) fits mid-size L&D teams producing 10+ courses per month. For enterprise (DLP, SSO, custom voices), reach out and we can discuss.
This is where AI voice absolutely wins. Set up a template script with placeholder stats (e.g., [YEAR], [PERCENTAGE]). Each annual refresh: update the placeholders, batch-regenerate the changed slides, swap the files. A 2-hour compliance course that would cost $800+ in re-recording costs is now a 20-minute script update. Keep all original scripts in version control.
Yes and it can actually improve accessibility. WCAG 2.1 AA does not differentiate AI vs human narration. What matters: you still need captions and transcripts (same as always). FreeTTS-generated scripts double as auto-captions in Storyline. Some learners with dyslexia or visual impairments find AI voice easier to control (speed, repeat) than a standard human recording.
Four do real work. <break time='600ms'/> for pause before a key concept. <prosody rate='slow'> for emphasizing a complex step. <say-as interpret-as='date'> for dates that otherwise sound robotic. <phoneme alphabet='ipa' ph='...'>word</phoneme> for technical terms with unusual pronunciation. FreeTTS supports all four. Test with a 60-second sample before generating the full course.
Sources
Where the recommendations come from
SCORM compatibility
SCORM 1.2 / 2004 and xAPI compatibility verified against scorm.com and the cmi5 spec for newer LMSs.