Cognitive accessibility, Easy Read and audio: what the voice fixes, and what it doesn't

When people talk about digital accessibility, they think first of sight and movement: colour contrast, keyboard navigation, compatibility with a screen reader. That is essential, and it is the heart of the standards. But there is a family of needs that rarely gets named in the same breath: cognitive accessibility, for people whose barrier is not seeing the text but decoding or understanding it. Learning disabilities, intellectual disability, age-related reading difficulty, imperfect command of the language: the text is right there, legible on screen, and still closed. Audio has a real role to play here, provided you say precisely which one, and do not pretend it fixes everything.

Two different barriers, often confused

You have to separate two things that get mixed up easily, because they call for opposite answers.

The first barrier is decoding: turning written signs into words. That is the effort a dyslexic person makes, or someone who reads slowly and tires. The meaning of the text is available to them if it arrives through a channel other than the eye. That is exactly what the voice does: it decodes on the reader's behalf. For this audience, a good-quality audio version genuinely changes access, which is what our pieces on audio and dyslexia or audio and attention difficulties go into.

The second barrier is comprehension: the sentence is too long, the vocabulary too abstract, the structure too dense. There, reading the text aloud is not enough. A forty-word bureaucratic sentence stays a forty-word bureaucratic sentence, whether you take it in with your eyes or your ears. Speaking a complicated text does not simplify it; sometimes, by unrolling it with no way for the eye to go back, it makes it harder to follow.

What Easy Read is, and why audio does not replace it

Easy-to-Read (also called Easy Read) answers the second barrier. It is a structured method, built on European guidelines led by Inclusion Europe, that rewrites the information: one idea per sentence, common words, the active voice, no metaphor, a concrete example where possible, and validation by the people it is meant for. It is not a summary, it is a rewrite in service of understanding.

An audio text and an Easy Read text therefore do not target the same need, and it matters not to confuse them. Audio takes a text and says it; Easy Read takes information and reformulates it. A text-to-speech tool does not produce Easy Read, and it should not be sold as if it did: that would promise a cognitive accessibility it does not deliver. The soundest form is often the combination: content rewritten in Easy Read, then also offered as audio, so the reader who needs both answers finds both.

Where audio genuinely helps cognitive accessibility

That said, the voice brings concrete things to a cognitive audience, beyond decoding alone.

• It lowers the reading load: following a voice takes less sustained effort than deciphering a screen, which helps people who tire fast or lose the thread.
• It enables the dual channel: reading and listening at once, with the text highlighting along with the voice, keeps many profiles anchored in the text.
• It hands over control of pace: being able to slow the reading down, go back, pause, makes the text something the receiver can steer.

The quality of the reading then counts twice over. A voice that stumbles on numbers, mangles acronyms or chops sentences adds load instead of removing it. That is why the finish of reading French, described on the French text to speech page, is not a luxury for this audience: it is the condition for audio to relieve rather than clutter.

What to keep in mind before you claim accessibility

Audio is a real lever for cognitive accessibility on everything to do with decoding and reading effort, and a useful complement for the rest. On its own it is not an answer to the complexity of language: that answer is Easy Read, and it takes an editorial effort a machine does not make. Announcing an audio version is honest; announcing full cognitive accessibility because you added audio is not. The right stance, the one we hold on our accessibility page, is to say what each measure fixes, and not to let a reader believe that a listen button opens a door that, for them, stays shut.

Read next

• Hyphenated compounds: the quiet trap for French text to speech
• Audio search: find the exact second a word is spoken
• SRT and VTT subtitles: one more file, not one more invoice