One of the changes in macOS Catalina was the removal of the “Enhanced Dictation” feature which allowed you to dictate offline in macOS Mojave.
This means that by default in Catalina, anything you dictate is sent to Apple for online translation by the Siri engine – which isn’t ideal if you’re dictating confidential documents or want to protect your privacy.
The good news is that you can still dictate offline in Catalina but Apple has changed the way you activate it.
You May Also Like:
In Mojave, the Dictation preferences look like below – you simply go to System Preferences > Keyboard > Dictation and check the box Enhanced Dictation in order for macOS to download the language packs necessary to dictate offline.
In macOS Catalina, the Dictation preferences look like the panel below – the Use Enhanced Dictation option has been removed.
In Catalina, there’s also no apparent way of configuring useful formatting commands such as “new paragraph” and “left indent”. Dictation in Catalina will also turn itself off after a few seconds of no voice input whereas in Mojave, it stays active until you turn it off yourself.
However the good news is that you can activate offline dictation in Catalina and still have formatting commands by going to System Preferences > Accessibility > Voice Control and checking the box Enable Voice Control.
This will then automatically download the language packs you need to dictate offline in macOS Catalina.
The English language pack is around 422MB in size so you’ll need to make sure you have enough space on your mac hard drive first.
If you click on the Commands… button, you’ll also see all the different formatting voice commands possible too that used to be available in Enhanced Dictation in Mojave.
You can also activate and deactivate Voice Control by clicking on the floating Voice Control Microphone window or saying “Wake up” or “Go to sleep”.
What’s Voice Recognition Like in macOS Catalina?
The voice recognition feature in macOS has never been brilliant, although it’s not terrible either. The main problem is that it never learns to adapt or recognize your voice or vocabulary like professional dictation software.
The other issue is that whilst it uses Siri for US English which does a pretty good job of transcribing, it still uses the same inferior Enhanced Dictation engine for other languages (such as UK English) for which the results are much poorer.
One of the main problems is there’s quite a delay in non US English languages between the time you speak and the time it takes to transcribe.
The recognition accuracy is also far worse and our advice to users in the UK is to select US English as your language of choice as you’ll get far better results than selecting UK English.
In fact the US English dictation accuracy in macOS Catalina gives Dragon Dictate a run for it’s money although it’s not quite on the same level yet.
Voice Control of your Mac meanwhile works much better in Catalina than dictating a document and if you just need Voice Recognition for this purpose, it’s faster and more accurate at controlling your Mac than previous versions of macOS.
Our advice is, if you haven’t upgraded to macOS Catalina yet and rely on dictation for transcribing documents, it’s worth upgrading if you just need it for US English.
For all other languages, it’s best to hold off until Apple has Enabled Enhanced Dictation in Catalina.
You can also let Apple what you think about dictation via the Apple macOS feedback page.