Dragon Naturally Speaking is available for PC and Macs. Which version is better?
I realize that this article is going to be a moot point for a lot of people. You’re either a Mac person or a PC person, and you’re naturally going to gravitate towards the version of the software designed for your computer, right?
Not necessarily. It turns out when it comes to deciding between the PC and the Mac version of Dragon Naturally Speaking there are several different options available to you depending on what kind of performance you demand. This is also a subject that’s been on my mind a lot lately as I’ve been switching my writing workflow to 100% dictation and I try to figure out a decent way of getting text from my recorder to a word processor.
Jun 16, 2016 The most seamless way to run the Windows version of Dragon on your Mac is with a virtual machine in Parallels Desktop. Unfortunately, the new Voice Control in Catalina is a big disappointment from a dictation standpoint and does not compare to Dragon in accuracy or customisability.
The problem is pretty simple. I made the switch to Mac about a year ago and I love it. This is coming from a lifelong PC user who grew up on the things. Seriously, my first PC was a monochrome 8086 IBM compatible that my dad spent thousands of 1980s dollars to buy. The only problem with that switch is I still like to use Dragon occasionally, and Dragon for Mac sucks.
Training: This is by far the best reason to get Dragon for PC. If it gets a word wrong then you can correct it and Dragon learns from that correction. If it repeatedly gets a word wrong then you can train it on that word and the problem goes away.
Seriously. I can’t tell you how much of a lifesaver this feature is. I’ve taught my Dragon how to swear. I’ve had it learn specialized vocabulary for fantasy and science fiction stories I was working on. It’s a game changer, a productivity saver, and something that you absolutely need in my opinion.
Dragon for Mac? Not so much. You just can’t train it the same way you can the PC version. I’m not sure what’s going on under the hood that they weren’t able to include the central feature of every PC version of this program going back to its inception, but it was a really boneheaded move. Dragon for Mac is basically a nice way to get your words on screen, but you’ll constantly be correcting the same transcription errors and It. Gets. Old.
Transcription is a mixed back between PC and Mac, but it’s a mixed bag that I think leans towards the Windows version even though there is a minor annoyance about the Windows version.
Transcription is how I use Dragon. I have a Philips recorder that I carry with me at all times so that I can utilize my downtime. If I’m on a drive then I’m dictating. If I’m in the parking lot waiting on my wife to do some shopping I can pull out the recorder and dictate. It’s a great tool for getting out a first draft and putting thoughts on the page, and because of that transcription is the thing I focus on the most when I’m setting up Dragon.
The nice thing about transcription in Dragon for Mac is that I can load up multiple files at once and tell Dragon to transcribe them, and then they’re transcribed in the background leaving me free to do other things. Compare this to the PC version of Dragon where you can transcribe multiple files at once by selecting them, sure, but the drawback is Dragon takes control of your PC while it’s doing the transcribing rather than doing that transcription in the background.
So it’s a game of tradeoffs. Dragon for Mac does the transcribing in the background, but remember that training I was talking about in my first point? Yeah, you really can’t do that with transcription. In Dragon for PC you create a separate input for your digital recorder under your existing profile and then you can train that input source as it makes mistakes and it will get better and learn how you talk.
Dragon for Mac? Not so much. It does transcription, sure, but it’s the same old problem where it’s going to keep making the same mistakes over and over again because you can’t train it so it never learns. The end result is you’re going to be spending a hell of a lot of time going back and fixing the same mistakes over and over again and believe you me that gets very old very fast.
So for transcription the tip of the hat goes to Dragon for PC.
Finally there’s accuracy to think of. How good are these programs out of the box?
I can remember a time in the late ’90s and early ’00s when you had to spend a lot of time training Dragon if you wanted anything approaching accuracy, and even then you still had to go over everything you wrote with an editor’s eye to make sure it was coming out correct. This was fine for my dad because he was a lawyer and lawyers employ secretaries to do dictation anyway. Dragon just made life easier for everyone involved.
But what about for an author who doesn’t have a secretary to go over everything? And that’s the rub of it. I’ve discovered that no matter what you do, no matter what version of Dragon comes out, there’s nothing that’s going to be one hundred percent accurate whether you’re talking about transcription or dictating to the computer. There’s always going to be little mistakes that creep in, and you’re always going to have to keep an eye out for those mistakes.
I’ve not done anything approaching a scientific study of this, but I have a general feel between using Dragon for PC and Dragon for Mac, and I’d say that for sitting down and dictating or for transcribing the accuracy is definitely better on the PC version out of the box. And since you can’t really do any training worth the name in Dragon for Mac it’s not like it’s going to get better, whereas in the PC version you can train and it’s going to do a better job of learning your unique style.
Dragon for Mac also has odd idiosyncrasies. The transcription sucks, as I mentioned, but it also capitalizes words randomly and inserts random spaces. There are a lot of little niggling details it gets wrong that adds up to a very frustrating experience for a piece of software that costs so much.
This is simple. If you have a PC then you need to get Dragon for PC. If you have a Mac? You still want to get Dragon for PC.
Stay with me for a moment here, because this is the solution I ultimately came up with since Dragon for PC is the one piece of software that I found myself missing when I made the switch to Mac.
Dragon for Mac costs $300. That’s a steep pricetag for a piece of software that’s essentially a less functional version of its PC counterpart. This is one piece of software where you’re definitely paying the Mac tax.
But don’t forget about Parallels.
The wonderful thing about today’s Macs is they’re fully capable of running a modern Windows OS, and it’s never been easier to run a virtual machine like Parallels that allows you to run a Windows install within whatever version of MacOs you’re running. Which means you get all the benefits of the one or two Windows programs you need to run while also retaining all your Mac stuff.
The cost makes sense too. Dragon Premium 13 costs roughly $120. Parallels costs $80 to either buy outright or to get a one year SAAS subscription that includes updates. That means you’re only out $200 to get Dragon working on your Mac, which is still $100 cheaper than buying Dragon for Mac outright! You don’t even have to worry about Windows, because Microsoft is giving away Windows 10 right now. The only penalty for not paying for Windows 10 is you get some annoying text in the bottom right corner of your screen and you can’t personalize the background, but why would you want to do that when you’re doing most of your computing on your Mac?
There are two potential drawbacks to this approach:
So there you have it. Avoid Dragon for Mac. Get Dragon Premium 13 for PC. If you’re using a Mac then you need to either run Dragon for PC in Parallels or install Windows on your system using Bootcamp and use Dragon for PC if you’re serious about voice recognition as part of your writing workflow.
That’s it for this update. Up next: Why Dragon isn’t the magic productivity silver bullet some people make it out to be, and why it can still be damn useful.