Making Course Material Accessible Without a Studio

You don't need a studio, a budget, or a semester of lead time to give your course an audio version. You need to pick the right material first, clean it up, and generate it in a way that doesn't create more work for you every week. This is a workflow for an instructor or a small learning team doing this alongside everything else you already teach.

It also isn't the whole solution to accessibility, and it's worth saying that up front rather than at the end. An audio version of your reading supports students who read slowly, who are learning in a second language, or who listen on a commute. It does not replace an accessibility audit, and it isn't a screen reader — more on that below.

Not everything in your course is worth narrating, and trying to do all of it at once is how this project stalls in week one. Readings and lecture notes are the highest-value place to start: they're already written in full sentences, students already spend the most time on them, and they translate to audio with almost no editing.

A document built for reading on screen carries clutter that only matters visually. Strip it out before you generate, because the reader hears every word you leave in, including the ones meant for navigation.

Pick a single voice and use it for every reading in the course, from week one to the final. Switching narrators partway through means students spend a few minutes re-adjusting to a new voice instead of listening to the content — a cost you pay every time you swap, for no benefit to anyone.

Audition a full paragraph, not one sentence, before you commit. A voice that sounds fine reading a title can turn tiring across a fifteen-minute reading, and that's the length you'll actually be generating most weeks.

Generate each reading or lecture note as its own file rather than pasting a whole module in at once. If a student flags a mispronounced term or you catch a typo in week six, you regenerate one short file instead of a forty-minute one.

This also matches how students actually use the audio: they play the reading for this week, not a bundle of everything up to this point. Smaller files load faster and are easier to skip around in.

Generation is priced in points, roughly one point per second of audio, with the exact cost calculated from character count rather than duration. A free account starts with 60 points, enough to test the workflow on one short reading before committing to anything.

For an ongoing course, Lite at $9/mo covers around 50 minutes of audio a month, which is workable for a handful of short readings a week. If you're narrating full lecture notes across several courses, Pro at $24/mo raises that to around 200 minutes and adds WAV export, which holds up better than MP3 if you're archiving audio across a semester. Full details are on the pricing page.

Keep the original document available every time you post an audio version. Some students want to read along with the audio, some want to search the text for a specific term, and some will only ever use the audio. Offering both costs you nothing extra once the file is generated.

Here are six voices that work well across different kinds of course content, from formal policy readings to conversational orientation material. Listen for the one that won't tire students out across a full semester.