Recently I took a tumble and fractured my right elbow, as well as spraining both my wrists. Although I am now becoming quite handy at left-handed typing and writing, at first anything like that was painful. So I decided to give the Speech Recognition software that comes with Windows 10 a try. Although it’s becoming more and more accurate over time, I wish I’d had some good tips on how to use it right from the start.

Over time, I’ve been discovering some links to useful instructions, and have also jotted down some tips of my own. If you’re interested in trying out the Windows 10 Word Recognition software, you might find the following tips and links helpful … and I’d recommend studying them before you charge into giving the software a try, as you’ll probably save yourself a fair amount of frustration.
1. Be very patient. Be very, very patient. There is initially a steep learning curve, and it can be frustrating. It is better to spend a few minutes a day over a month or so than to get frustrated and give up.
2. Here is a very good guide to actually setting up your Windows 10 Voice Recognition software and getting it accustomed to your voice.
3. You will get stuck from time to time, so here is the link to the help for Windows speech recognition:  As well, download this great PDF that goes into detail on what is required to reach a 99% accuracy with your dictating—excellent tips and advice! Although it was written for Windows 7 and/or Vista, the vast majority of the information is viable for Windows 10 as well.
4. Make sure you are in a very quiet environment. Even white noise, such as an air conditioner, can affect how it works. Very strange results can come up if someone comes in the room and you have a conversation with them, without first telling the software to “Stop listening.”
5. You must enunciate very slowly and clearly. Be very clear with vowels. Unfortunately, the Windows speech recognition software seems to also get confused by sets of letters such as v and b, k and p, f and th and sh, etc. So when saying words that start with those kind of letters or sounds, be sure to say them slowly and clearly. Be careful not to run words together, but at the same time, make sure you say multi-syllabic words as a whole.
6. Be sure, when you start, to make good use of the practice phrases and paragraphs. Use all the provided  sets and use each one at least 2 or 3 times. This same link will also teach you how to correct mistakes.
7. Think carefully before you talk, as it works a lot better if you have a full sentence ready to say. If you only say one word or a phrase it is more likely to make mistakes. Think of the sentence before you start, because if you hem or haw it will get confused. It also works quite well if you break a long sentence into its parts, stopping at commas or colons.
8. The longer you work on it, the better it works. It takes a lot of practice, until it understands you reasonably clearly.
9. There are sheets you can print out with key words you say to give different commands, but you are better off to pick out the commands you think you will use, and make yourself a one-page sheet listing your often used commands, which you can refer to easily and quickly.
10. When using the feature that corrects spelling, take a careful look through the entire list of all options they present, instead of just saying the word again if you don’t see it right away. Once the correct word has been suggested, they won’t give it again as an option.
11. If you tried to make the correction two or three times, and it didn’t work, you are better off to just manually correct it. It seems like there are certain words it never gets correct, such as, in my case, “again,” which the software never gets right no matter how I say it. I think you have to accept that the voice recognition software is meant as a help, but not as a total solution to getting out of typing. If you want voice recognition software that works better, I have been told that Dragon is the way to go.  On the other hand, there are other speech recognition options as well, such as Google’s; this post discusses how the Windows 10 version compares to other choices.
12. A funny thing is that our voices change as the day goes by. As we get tired, we tend to slur even though we don’t realize it. For best effect, do your dictating when you are fairly wide awake and alert!
13. By the way, you can also train the “Cortana” feature in Windows 10 to recognize your voice and follow your voice commands. Here is a good guide.
Have you used Windows 10 Voice Recognition software or some other brand? What tips can you share with us? Please share your insights in the comments below. Thanks!