Ask a learning support teacher what they were advised to do about dysgraphia fifteen years ago and you’ll hear the same answer: put the student on Dragon.

It was sensible advice. It just didn’t work for the students who needed it most.

Dragon was trained on adult voices — adult pitch, adult pace, adult vocabulary, adult articulation. Put an eleven-year-old in front of it and accuracy fell apart. The child would speak a sentence, watch the screen produce something that wasn’t remotely what they said, and try again. And again. Eventually a teacher would quietly shelve the software and go back to a scribe or extra time, and the student went back to fighting the pen.

That was the state of things for a long time: the accommodation that should have been transformative for primary and early secondary students was effectively age-gated, and nobody said so out loud.

This is what has changed. Speech Recognition Cloud is built to recognise younger, still-developing voices accurately — so dictation is now a realistic accommodation for a nine-year-old with dysgraphia, not just a sixteen-year-old. For a learning difference that shows up in Year 2 and does its worst damage long before senior school, that difference is not a minor upgrade. It is the whole point.

Dysgraphia is not bad handwriting

The most common misunderstanding about dysgraphia is that it’s a neatness problem. It isn’t.

Writing by hand requires several systems to run at once: visuospatial processing to place letters, graphomotor control to form them, working memory to hold the sentence you’re partway through, plus spelling, grammar and punctuation. In a fluent writer, most of that is automatic and costs nothing. In a student with dysgraphia, one or more of those systems demands conscious effort — and effort spent on letter formation is effort not available for the idea.

This is why the work a student with dysgraphia hands in so often understates them. A child who can explain the causes of the Second World War in detail out loud produces four laboured sentences on paper, and the four sentences are what gets marked. The gap between what they know and what they can evidence isn’t a comprehension gap. It’s a transcription gap.

Handwriting practice doesn’t close it. Typing helps some students and not others, because typing still demands spelling, spacing and motor sequencing in parallel with composition. Dictation is different in kind: it removes the transcription load entirely and lets the student spend everything they have on the thinking.

What actually changes when a student can dictate

Three things, consistently.

The output finally matches the ability. Students stop being marked on their weakest channel. Teachers frequently report being surprised by the first dictated assignment — not because the student improved, but because they’d never seen what the student could actually do.

The dread lifts. When every writing task has historically been slow, effortful and disappointing despite real effort, avoidance is a rational response. Homework battles, “I forgot it,” the ten minutes of sharpening pencils before starting — that’s usually not defiance, it’s anticipated failure. Dictation lowers the cost of beginning, and the avoidance tends to fade with it.

Length and complexity go up. Students with dysgraphia learn to write short, because short is survivable. Given a channel that isn’t punishing, they write longer sentences, use the vocabulary they actually have rather than the words they can reliably spell, and revise — because revising no longer means writing the whole thing out again.

The research backs this up, with one important caveat

A 2025 review in Disability and Rehabilitation: Assistive Technology examined five years of studies on assistive technology for students with dysgraphia. Speech-to-text came out as one of the most widely used and effective supports, alongside word prediction and graphic organisers, with consistent gains in academic performance and classroom engagement where it reduced reliance on manual writing.

The caveat matters as much as the finding: outcomes are strongest when the tool is matched to the individual student. Dysgraphia isn’t one profile. Some students are held up mainly by the motor act of handwriting; some by spelling; some by organising thought into written structure. Dictation solves the first two decisively. For the third, it works best paired with planning support — an outline, a graphic organiser, or simply a teacher asking the student to talk through their argument before they dictate it.

For context on scale: writing-related learning differences are estimated to affect somewhere between 5% and 20% of students, and language-based learning disabilities as a group affect an estimated 15–20% of the population, according to the International Dyslexia Association. In a class of thirty, this is not an edge case.

Making it work in practice

A few things separate the students for whom dictation sticks from the ones who abandon it in a fortnight.

Frame it as an accommodation, not a shortcut. Glasses aren’t cheating at reading. Students pick up adult ambivalence quickly, and a child who suspects they’re getting away with something will stop using the tool the moment a peer notices. Say plainly: this is how you write now.

Teach the speaking. Dictating well is a skill — full sentences, punctuation commands, a beat between ideas rather than trailing “um.” Nobody is good at it in the first session. Budget two or three low-stakes goes before it’s used on anything that’s assessed. Our guide to spoken punctuation is a good starting point.

Get the microphone right. Recognition accuracy depends heavily on clean audio, and a laptop’s built-in mic in a room with twenty-nine other children is not clean audio. A decent headset is the single cheapest accuracy improvement available.

Don’t retire handwriting entirely. Most students still benefit from handwriting and typing practice in low-stakes contexts. Reserve dictation for the work where the ideas are the point: extended writing, timed assessment, anything long enough that transcription load becomes the limiting factor.

Where Speech Recognition Cloud comes in

Speech Recognition Cloud was built by Russell Busill, who has spent almost 27 years working with Australian schools, students and people across a wide range of disabilities and support needs.

That’s a long time watching the same failure repeat: a capable student handed software that couldn’t hear them properly, then quietly written off as someone dictation “doesn’t suit.” It suited them fine. The tool didn’t fit the user — and for younger students especially, nobody was building for that user, because the market was adult professionals.

Speech Recognition Cloud was designed from that starting point rather than adapted toward it. You put the cursor where you want text, press a hotkey, speak naturally, and the full punctuated paragraph appears — in Word, in Google Docs, in an online assignment form, anywhere on Windows. It sits within the same speech to text for students range from Voice Recognition Australia used in schools around the country, and there are dedicated pages on dictation for students and speech to text in the classroom if you want the detail for a specific setting.

The honest way to evaluate it is not to read about it. Put it in front of a student on real work — a piece of homework, a draft they’ve been avoiding. The free edition includes 20 minutes of dictation a month, which is more than enough to watch a child dictate a paragraph and see whether the screen shows what they actually said.

For a lot of families, that first paragraph is the moment it becomes obvious the problem was never the child.

Sources