Is Voice Recognition a Viable Alternative?

 For the past couple of years, I have been using voice recognition software more frequently. Some of you may have noticed my use because of the insertion of random words that seem to make no sense. As I have written about previously, my use of voice recognition goes back to the introduction of primitive system for the IBM Selectric typewriter back in the 1960s. I probably tried and purchased almost every program available. From time to time, I have written posts about my experience in using voice recognition.

Just recently, I purchased another update to the Dragon Dictate program from Nuance software. This is the Macintosh version of the Windows program called "Dragon NaturallySpeaking." To give you an idea of the accuracy of voice-recognition I am going to dictate the remaining portions of this post without making anything except voice commands. In other words, if the program makes a mistake in transcribing my voice, I am going to leave those mistakes in the post.

 I am painfully aware that smart phones and tablets have implemented primitive voice recognition. The most annoying of these is the Apple Siri program. But Google has also implemented a rather limited voice recognition option and voice recognition is showing up in many other places. The question from my standpoint is not whether I can get my smart phone, in my case an iPhone, to respond to my voice commands but whether or not I can use voice recognition as part of my normal activities of writing huge amounts of text. As I've often noted in the past, the issue is accuracy. If I have to spend as much time editing the text as I spend writing it, I might as well type the text and make the corrections as I go along. As a side note, I note that one of the skills for which I am often cited in the LinkedIn website is editing. If you are a daily reader of my blog you probably realizing that editing is not necessarily my greatest talent.

I am also aware that both Windows and the Mac operating system have built-in voice recognition software. I would view both of these programs as rudimentary. For text that requires a great deal of formatting, voice-recognition is not much of a help unless you are physically limited in your ability to enter text by a keyboard. I must admit that I do not use voice recognition to control my computer, primarily because I am so much faster with a trackpad.

I am always surprised at the accuracy of voice recognition programs in recognizing names. But this does not mean that I would use them to enter data into a genealogical database program. For example here is a list of names of some of my ancestors:

• Henry Martin Tanner [correct]
• Ovi Christian over some [Ove Christian Overson]
• David Thomas [correct]
• Jens Christiansen [Jens Christensen]
• Carol Morgan [Harold Morgan]

You can probably see, that I would spend as much time correcting the entries as I would entering them.

One thing I have learned, is that it is important to read the text that you dictate and in effect, do your editing at the time that you are reading the text. It is also important to take the time to train the programs.

Actually, I am pretty much impressed with the increased accuracy of the Dragon Dictate program. Because the program has been constantly improved, I have been willing to spend for the cost of the upgrades. The point at which voice-recognition software became a viable option was when its accuracy became essentially the same as my accuracy while typing. I must admit, that it does a much better job of spelling words the first time that I do. Without the built-in spellchecker in my computer, my spelling would be much worse than it appears to be.

Well this is the end of the post that has been entirely dictated with the Dragon Dictate version of the program for Macintosh OS X operating system. Oh, I almost forgot, voice recognition didn't even begin to work until computers got fast enough to process vast amounts of information in a very short period of time. If you are using an older computer you may not experience the full capabilities of the programs.