macOS finally lets you record custom vocabulary in Voice Control – but Apple still has work to do
Did this really just happen?
Tucked away in a recent macOS Sequoia update, Apple has quietly rolled out a long-awaited feature: you can now record custom vocabulary in Voice Control on the Mac.
This feature first arrived on iOS in version 18.0 last year — now, at last, macOS has caught up. No fanfare, no splashy announcement.
As one prominent Apple influencer told me, “It’s quite crazy Apple didn’t tell us.” But for disabled people who rely on voice-first computing, it’s a big deal.
So what’s changed, why does it matter — and where does Apple still fall short?
What is custom vocabulary in Voice Control?
Voice Control is Apple’s built-in speech recognition system that lets users navigate and dictate across macOS and iOS entirely by voice. Until now, you could create custom commands, but you couldn’t train the system to recognise your own names, phrases, or specialised vocabulary with recordings that reflect your voice.
That’s exactly what this new feature enables.
You can now add a custom word or phrase and record yourself pronouncing it, helping Voice Control better understand your speech — particularly useful for:
- Personal names or place names
- Accents or non-standard pronunciation
- Medical, legal, or technical terms
- Phrases that Voice Control often mishears
👉 For step-by-step setup instructions, see Apple’s official support guide on using custom vocabulary.
Why it matters
This change improves more than just transcription accuracy. It supports:
- Accessibility: For disabled people with speech impairments or strong accents, being understood by technology is not a luxury — it’s a necessity.
- Productivity: No more workarounds to spell out difficult phrases or repeat words until they’re recognised.
- Personalisation: Custom vocabulary makes workflows smoother and better adapted to each individual user.
- Inclusion: Technology that adapts to people — not the other way around — is fundamental to inclusive design.
As someone who has long campaigned for this feature, it’s heartening to see Apple finally deliver. But for Voice Control to be truly transformative, this needs to be the start of a much deeper investment in the voice app— not the end.
What’s missing: buried features and broken expectations
- Re-recording is hidden:: Yes, you can re-record a pronunciation — but only by Control-clicking on the entry. There’s no visible button or prompt. For an accessibility feature, that’s poor usability.
- No adaptive learning: There’s still no sign that Voice Control improves its recognition over time, unlike Dragon, which adapts to corrections and gets smarter the more you use it. Apple’s system remains static.
- No voice-only vocabulary editing: Ironically, adding new words still requires navigating menus manually. There’s no hands-free command like “add to vocabulary” followed by a recording prompt. For disabled users who can’t use a keyboard or mouse, this remains a barrier.
👉 For a deeper comparison, see our review of the top voice dictation tools for disabled people in 2024.
Final thoughts
This may be a quiet update — but for Voice Control users, it’s a meaningful one. It shows that Apple is listening, at least in part, to the accessibility community’s call for greater flexibility, accuracy, and personalisation.
But it also shows how far there is still to go.
If Apple wants to match — or surpass — the sophistication of Dragon and other leading voice systems, it needs to build in adaptive learning, embrace hands-free workflows fully, and make its accessibility tools truly discoverable and intuitive.
Until then, this remains a step forward, but not the leap many of us are still waiting for.
Have you tried the new custom vocabulary feature in macOS Voice Control? Share your experience in the comments below or tag us on social media.
