Reading aloud to a camera is hard work for a lot of people, and most teleprompters make it harder: they set the pace, they punish a lost place, and they treat the typeface and the highlight as decoration. TellyPrompter was built the other way up. This page describes, concretely, what it does for anyone who finds reading aloud difficult — including dyslexic readers — not as a settings menu bolted on afterwards, but as decisions in the core of the product.
A scoping note first, because this page should earn trust rather than assume it: what follows describes what the product does, not outcomes for any group of readers. Reading difficulty is individual, and the only test that matters is reading your own script in it, which is free and needs no account.
OpenDyslexic is a script face, not a toggle
TellyPrompter ships OpenDyslexic as a first-class script face. That phrase is doing real work: it gets the same size range and the same reading states — the highlight, the active line, all of it — as the default face. It is chosen where you choose any font, not buried in an accessibility panel, and nothing about the reading experience is a reduced version when it is on.
Whether OpenDyslexic helps you personally is something only you can say — readers differ on it, honestly and famously. The point is that if it is your face, the product treats it as a first choice rather than an accommodation.
Your place is never carried by colour alone
The single most load-bearing element in a teleprompter is the mark that says you are here. Lose it and you have lost your place mid-sentence, on camera. So in TellyPrompter, no reading cue is carried by colour alone — position, weight and form do the work too, because roughly eight per cent of men cannot reliably resolve a colour-only distinction, and a highlight you cannot see is a prompter that does not work.
The same thinking sets the contrast. The product assumes it may be read through beam-splitter glass that has already taken a third to a half of the light, so nothing critical depends on a subtle tonal step. If your rig is just a phone on a desk, that headroom is still yours.
The scroll follows you, which changes who reading is hard for
A conventional prompter asks you to read at its speed. If reading aloud costs you more effort on some words than others — which is what reading difficulty is — a fixed scroll turns every hard word into a race, because the text keeps moving while you work.
TellyPrompter's voice mode follows instead: the highlight tracks the word you are actually saying, and when you slow down, stop, back up or re-say a line, it waits and picks you up. The pace of the take is whatever your pace is, sentence by sentence. Speech recognition runs on the device and the audio is never uploaded. The honest limits: voice mode relies on the browser's speech recognition, which Firefox does not provide — the constant and sound-level modes work there — and recognition quality is the browser's, so it varies by platform, language and microphone.
Alongside it, the practical companions: tap any word to jump there, a settable active-line position so the reading line sits where your eyes want it, look-ahead you control, and stop markers at sentence punctuation.
The rest of the reading surface is yours to set
Typography and spacing, text size well beyond timid defaults, colour themes, alignment, and full text direction — left-to-right and right-to-left, across 34 languages with automatic detection, including non-Latin scripts. Reduced motion is honoured: if your system asks for it, animation collapses to instant. And mirror mode is a proper whole-surface flip for prompter glass, so nobody is asked to read reversed text.
None of this is presented as a medical claim, because it is not one. It is a reading surface built on the assumption that readers differ, at every point where a prompter usually assumes they do not.
What this page deliberately does not say
You will notice this page does not claim a compliance standard. Our conformance target for the site and app is WCAG 2.2 Level AA, and it stays described as a target until complete user journeys have been audited — describing tested behaviour honestly seems more useful to you than a badge. And it does not claim that any tool removes the difficulty of reading aloud; a good one just stops adding to it.
If reading aloud to a camera is the thing you have been avoiding, the gentlest test is the real one: paste a few sentences, pick the face and size that suit you, and read them once with the scroll following you. The guide to reading naturally picks up from there, the language piece covers the 34 languages properly, and the free-tier article explains why all of the above costs nothing.