Posey's Tips & Tricks

Taking a Fresh Look at Dictation in Microsoft Office

Be careful what you say -- and how slowly you say it.

As someone who writes an insane amount each month, I have long relied on dictation software to help me get my assignments done on time.

I first started experimenting with dictation back in the mid-1990s, using an IBM application called ViaVoice. Unfortunately, at the time ViaVoice wasn't really suitable for professional use. I found that its discreet speech requirement and limited vocabulary made me write more slowly than if I had simply typed the assignment.

For most of my writing career, I have been using Dragon NaturallySpeaking from Nuance. I currently use Version 15, and the software's speed and accuracy is nearly flawless. Sure, I have taken the time to tune the software in an effort to achieve the best possible accuracy, but my efforts have not been in vein.

As fantastic as the current version of Dragon NaturallySpeaking may be, however, it isn't perfect. It's not that the software is buggy or anything like that. (Nuance seems to have ironed out the bugs a couple of versions back.) Instead, there are little things that can sometimes get in the way of optimal efficiency. These are small things that are usually related to formatting.

For example, Dragon might spell out a number instead of typing a numerical value, or it might write words in a way that contradicts a publisher's writing guidelines. For instance, Dragon might type the word "datacenter" as two words instead of one ("data center" instead of "datacenter").

In other cases, Dragon's unique nuances prevent me from being able to use it at all for writing certain types of columns. This is especially true for columns that cover a step-by-step process. The reason for this is that there are certain words that Dragon interprets as commands. The most problematic of these commands is the Select command. Let me give you an example.

Let's pretend for a moment that I am trying to tell someone how to save a Microsoft Word document. In doing so, I might use a phrase such as, "Select the Save command from the File menu." Dragon hears the word "select" at the beginning of the sentence and interprets it as a command rather than something to be typed. It therefore selects an instance of the next word that I speak, which, in the case of the example phrase, would be "the."

Once Dragon has selected a block of text (in this case, the word "the"), it overwrites the text with the next thing that I say. In this case, the phrase "Save command from the file menu" would be inserted into a random sentence earlier in the document, where the word "the" used to be.

Although I have contacted Nuance a few times over the years and tried to get them to create an option to disable the Select command, I haven't gotten anywhere with my efforts. I have come to accept the idea that Dragon NaturallySpeaking is a great tool, but I can't use it for everything that I write.

Of course, Dragon is not the only dictation tool in existence. Recently, I installed an update for Microsoft Office 2016 Professional Plus and noticed that the update placed a Dictate icon on the toolbar. You can see this icon in Figure 1.

[Click on image for larger view.] Figure 1: The Microsoft Word toolbar now contains a Dictate icon.

Microsoft has dabbled in speech recognition for quite some time. The Windows operating system is capable of recognizing speech, and Microsoft has offered dictation capabilities for Word through an add-on for quite some time. Over the years, I have noticed Cortana's speech recognition accuracy improve dramatically, so I was curious to see if Word's dictation capabilities were as good as (or possibly even a viable replacement for) Dragon.

What I have found is that the Office dictation engine does indeed accurately recognize speech. The dictation engine does sometimes recognize words incorrectly, but Word usually recognizes its own mistake and fixes the problem by the time that I finish speaking the sentence.

The one big problem that I have with Office's native dictation feature is punctuation. When I am dictating in Dragon, I can verbally insert punctuation. When I get to the end of a sentence, for example, I speak the word "period" and Dragon will insert a period. Dragon's punctuation works quite well.

In the case of Office, however, Word automatically inserts a period any time that I stop to collect my thoughts. The end result is that a sentence might be broken into two or three fragments, simply because I did not speak the rest of the sentence quickly enough. And if I spend too much time collecting my thoughts, Word will turn off my microphone.

To give you an example, here are two sentences that I dictated in two different ways -- first with Dragon, then with Word. In both cases, I deliberately paused after speaking the word "example."

  • This is an example of a sentence that was dictated using Dragon Naturally Speaking.
  • This is an example. Of a sentence that was dictated using Microsoft Word.

You can see the stray period and subsequent capital letter that Word has inserted.

My assessment is that if you speak very fluidly and know exactly what it is that you want to say, then the Microsoft dictation engine will probably work for you. Better still, unlike Dragon, the Microsoft dictation engine does not require any initial set up or training. All you have to do is click on the Dictate button and start talking.

For me, however, I find Dragon to be much more forgiving and far better-suited to the way that I work. I think that it's safe to say that I will be using Dragon for the foreseeable future.

About the Author

Brien Posey is a 22-time Microsoft MVP with decades of IT experience. As a freelance writer, Posey has written thousands of articles and contributed to several dozen books on a wide variety of IT topics. Prior to going freelance, Posey was a CIO for a national chain of hospitals and health care facilities. He has also served as a network administrator for some of the country's largest insurance companies and for the Department of Defense at Fort Knox. In addition to his continued work in IT, Posey has spent the last several years actively training as a commercial scientist-astronaut candidate in preparation to fly on a mission to study polar mesospheric clouds from space. You can follow his spaceflight training on his Web site.

Featured

comments powered by Disqus

Subscribe on YouTube