A voice-over for your courses and e-learning modules
Forty modules, three updates a quarter, two languages: narrating an online course is a job that never ends. An AI voice makes it consistent, editable and affordable.
Consistent narration across dozens of modules
The first problem with a course recorded by people is inconsistency: one module recorded in January, another in June, a trainer with a cold, a different microphone. With a synthetic voice every module comes out with the same timbre, the same pace and the same loudness, from chapter one to chapter forty. How the engine works is explained on the AI voice generator page; for a course, the key point is that it reads exactly what you write.
Update a slide without re-recording everything
A price changes, a procedure evolves, a screenshot is replaced: you edit the sentence concerned, regenerate that passage and swap the audio file on the slide. The rest of the module is untouched. One generation covers 1,500 characters, the narration of one dense slide or two or three short ones, which maps naturally onto how a course is built.
Bilingual courses: one script, two voices
For training delivered in English and French, translate the script, then generate each version with the matching voice: Paul, Oliver or Jane in English, American or British accent as you prefer, and Marie in French. Pace and quality are identical in both languages, something a human recording rarely guarantees. Compare the voices on the voices page, where every sample reads the same text.
Numbers, acronyms and technical terms
Teaching content is full of figures, dates and acronyms. Digits and dates are read correctly; common acronyms are pronounced the way people say them. For a rare industry term or a proper noun, test it on its own with the free trial, two sentences with no account, and adjust the phonetic spelling if needed before generating the whole module. The text to speech page goes into these writing adjustments.
Bringing the audio into your tools
Download the mp3 and import it into your authoring or presentation tool: Articulate Storyline or Rise, Genially, iSpring, PowerPoint (one audio file per slide), or DaVinci Resolve and Premiere for training videos. WAV is available if your production chain prefers an uncompressed format. Name files by module and slide; your history keeps them for 90 days (see how long audio is kept), long enough to finish the version.
A practical workflow: keep the script in a spreadsheet with one row per slide, generate the rows that changed, and replace only those files. Your scripts are never stored on our side, only a technical fingerprint, so the spreadsheet remains the single source of truth for the whole course.
Accessibility and transparency
Clear narration at a steady pace helps learners who struggle with reading or follow the course in a foreign language. Pair it with captions generated from your script, which is clean by construction. State in the module that the voice is synthetic: the European AI Act expects it when content could pass for a real person, and every mp3 already carries a technical marker in its metadata.
Budget for a two-hour course
Two hours of narration is about 120,000 characters, applying the counting rule of roughly a thousand characters per minute. That is one €9 pack of 100,000 characters plus a small top-up, or the Medium subscription at €14.99 a month for 200,000 characters if you produce a course every month. The pack stays valid for twelve months and commercial use is included with paid credits (see the FAQ on usage rights); all prices are on the pricing page. For shorter video capsules, the YouTube voice-over and AI voice for TikTok guides complement this one; for audio-only formats, see voice-over for podcasts.