Dragon Dictate for Mac breaks its ‘Golden Rule’ – and users rejoice

“Dragon Dictate for Mac, the only viable option for Mac speech recognition, turns 2.5 today. The new version, a free upgrade for Dictate 2.0 users, addresses one of the most common complaints about the program: you can’t use the mouse or keyboard while entering text with your voice,” Nate Anderson reports for Ars Technica. “Or rather, you can, but your document will probably suffer.”

“The problem was bad enough that Nuance, the developer behind Dictate, laid out a “Golden Rule” against ever using the mouse or keyboard while speaking to the program,” Anderson reports. “If you’ve ever used voice recognition, you know that on plenty of occasions it is simpler just to lean forward and tap out a complex name than to spell it letter by letter; Dictate made this a dicey proposition.”

Advertisement: Limited Time: Students, Parents and Faculty save up to $200 on a new Mac.

Anderson reports, “With version 2.5, Dictate finally ditches the Golden Rule, at least in compatible apps like Microsoft Word 2011. While the feature may sound trivial to non-users, Nuance notes that it ‘was the most requested product enhancement from Dragon Dictate customers.’ Mixing keyboard, mouse, and voice no longer disrupts the program’s internal document map.”

Much more in the full article here.


  1. If you are such a mushmouth that DD makes such obvious mistakes, then you’re better off spending your bucks on speech lessons, not speech recognition.
    Me and my “Golden Voice” just plain *love* Dragon Dictate!

    1. No @Paul I. Not all words dictated to a computer are common English words. In today’s ‘New World Order’, dictation includes words from countries all over the planet as well as seemingly unpronounceable names. And as @mdmac points out, not all people dictating in English use it as their first language, meaning that they have dialects. Then there are the numerous English dialects, the day you have a bad cold, the noisy dictation environment, the POS microphone you’re forced to use in the field, blahblahblah.

      Killing off the asinine ‘Golden Rule’ is a return to sanity.

      1. I also applaud the end of the “Golden Rule.”
        As an actual user of Dragon Dictate, I know that it trains itself to understand each specific user’s voice. So the comically-intended “mistakes” listed by others are, in fact, unlikely to occur.
        That was all I meant to convey in my original comment.

        1. I concur Paul, the noise from any source has to be understood. Take your typical moron, whose native tongue is a dialect only two-year-olds can understand, and it could take months for them to learn DragonDictate’s proper command dialect.

          In other words, articulation is to speech, what Dragon Dictate is to taking complete control of the Macintosh, using the power of your voice.

          Those two paragraphs, were dictated to DD, after which the words were copied to the Clipboard and pasted in the MDN comment window. I did not alter the dictated text other than to add HTML Italicize. Hands-free and it took little more than a minute.

          With practice, you could do it with your eyes closed. But if you talk all day with gum in your mouth, don’t spit it out as you train Dragon to understand your words.

          Unfortunately, DD’s system doesn’t support audio cues, which would could increase productivity ten-fold. For example, it would be nice is there was a swish when windows opened and a click when they closed. Custom audio cues would be even better. I could import my Warner Brothers audio library.

          But the coup de maître is that Dragon relies heavily on AppleScript for all Mac-related functions and these code snippets are highly customizable.

  2. Ironic that right here on this page is an ad for Dragon picturing an Indian woman speaking into a microphone. Hell, if this thing can understand her then I’ll buy it so I can understand what my utilities are saying when I call customer support!!

Reader Feedback

This site uses Akismet to reduce spam. Learn how your comment data is processed.