Write the lesson, pick a voice from OpenAI or Gemini, and put it over your screen recordings or slides in Studio — with subtitles burned in or exported as .srt. When the product changes, regenerate one line instead of re-recording a module.
1,000 premium characters to try, on the house. No card, no trial clock.
Up to 5,000 characters per generation, about five minutes of narration. Numbers, dates, units and abbreviations are rewritten into words first, so “3.5 kg” and “Q4” are read as a trainer would read them.
“Calm and clear, for beginners”, “brisk, for a product update”. One voice and one direction across a whole course keeps every module sounding like the same instructor.
Voice over a screen recording or a slide deck exported as images — up to twenty slides make one narrated video, timed to the voice. Up to eight voice clips can be placed on one video.
Burn subtitles into the video or download them as an .srt file for your LMS player. Export at 720p or 1080p.
Courses change after they ship. The narration should be the easy part to change.
A feature was renamed? Edit the line, regenerate it, drop it back in. No booking, no matching the old recording.
Paste the translated script and voice the same module in another of 59 languages; the edit stays the same.
Up to twenty images become a narrated video, each slide held for its share of the voice.
Burned in, or as a sidecar file — for accessibility and for learners watching without sound.
On paid plans, bulk generation voices up to 500 lines per submission and returns them as named files in one ZIP.
A REST API with webhooks for teams that build narration into their authoring pipeline.
Better to know before you sign up than after.
| ProsodyAI | |
|---|---|
| Translate the script | No — you bring the translation; the voices speak 59 languages |
| SCORM or xAPI packages | No — you export video, audio and .srt, then add them to your authoring tool or LMS |
| Your own recorded voice | No — you choose from the built-in OpenAI and Gemini voices |
| Long source videos | Up to 20 minutes per uploaded video on paid plans, 5 minutes on free |
Yes. Paid plans include the commercial licence, and the audio you generate is yours to use in paid courses and internal training.
A minute of narration is roughly 800–900 characters. Starter’s 100,000 premium characters a month is about two hours; Creator’s 400,000 about eight; Studio’s 1,000,000 about twenty. Editing in Studio does not use characters.
Yes. You pick a voice by name and it reads every lesson; module twelve sounds like module one. The direction changes how it reads, not who reads it.
It produces standard files — MP4 video, MP3 or WAV audio, and .srt subtitles — that any LMS or authoring tool accepts. It does not build SCORM packages.
Yes. A new account starts with a one-off 1,000 premium characters, no card — enough to hear a lesson’s opening on both engines. Studio is on every plan.
Paste a lesson, hear it on both engines, put it over your slides. 1,000 premium characters to try, no card.