We just hung a TouchSmart in the kitchen and it’s working out fantastically well—would a program like Dragon Naturally Speaking make things easier or more frustrating?
We’re not really interested in dictating anything, more running the computer. Is it overkill? Is there a smaller app that does the same thing but without punctuation?
Will it work with two people’s voices or does it have to be ‘trained’ to one voice?
Speaking of which, how easy is it to run a computer with it? Can it be done? Can I have it Open File Explorer … Navigate To… Open … blah blah. Is it really rudimentary or will we still need a mouse?
Same question about browsing. Can it handle simple or a bit more complex commands? Got to … Straightdope.com …
Wait. How do you click links?
You can clearly tell I’ve never seen it in action. Any thoughts?
(Oh, before it’s trained, will it go on papers?)
Thanks,
It works pretty well, but can be frustrating. How can it recognize “nissen fundoplication via endoscopy complicated by pneumoperitoneum” correctlyevery time but insist on writing “Qadgop the Mercotan” as Can God the Merciful, despite repeated training and correction?
I don’t use it for much more than dictating onto Word documents, then sending them on to be printed up and pasted in the chart.
My version is trained to my voice and I’m the only one to use it. Other physicians who dictate in our system have their own copy, trained to them. Each one costs a bundle, but saves time.
So that input doesn’t probably help you at all. Sorry.
I used it at one time in the past. Like all the speech recognition software I tried out I gave up as the errors were to annoying to deal with. My voice really couldn’t handle speaking all the time either. Software that is meant to recognize a few commands has a broad leeway built into the recognition pattern allowing most voices to work with it. The fewer words it has to differentiate from in it’s database the less errors it makes. The more words it has to recognize the less tolerance there is for variance to the pattern recognition and it must be tuned to one voice for something like voice to text dictation. You also have to spend a lot of time reading text to calibrate the software for dictation. Background noise has to be kept to a low noise ratio to your voice input too.
I tried it… but many, many years ago. I’m sure it’s much better these days.
Back then it wasn’t terrible, but it wasn’t quite good enough to use. I would assume that the passage of time, massively increased processing power, and software developments, must have improved it tenfold.
I use ten standard currently for “typing” documents, and I’ve had every version since five. Each is better than the last. Five was nearly useless (thank god I paid next to nothing for it on ebay years ago), but nine and ten are what people probably expect from a voice recognition program.
It works pretty well, but has the terrible tendency to insert whatever pronoun it feels like - even in a New England accent “I” and “she” sound nothing alike, so I have no idea why this persists still…or why it insists on capitalizing “windows” as if people are going to write about the OS more often than the things you look out of. There is also a handful of words that I must say strangely, like almost and calm, because while it’s about 95% accurate pronouns not withstanding, it never understands those words. It’s a lot quicker to use it and proofread than to type things out, so that outweighs the minor drawbacks.
And yes, every user must train it to his/her voice. It doesn’t take too long, though, two hours if you want to be really thorough, though you could run it within 45 minutes.
I used the Preferred Version on Windows 7. I use it for legal stuff. It is quicker than typing. I am sure as hell not going to spend another $1200 or whatever it is for the legal version, so I cannot add anything there. I’ve liked it since Version 9. I don’t use it for writing posts on teh Innertubes
Don’t use it myself, but I know David Pogue, the Tech reviewer for The New York Times, is a big fan. Here’s a video demonstration he did with Dragon Naturally Speaking 10.
Terry Pratchett uses Dragon Dictate, with an additional plugin called Talking Point, to dictate his book writing, now that his Alzheimer’s has affected his ability to use a keyboard.
Some of the lawyers and their assistants in our office use it, and seem to like it. My assistant would like me to use it, but I’d like to know more about it. (And threads like this will certainly help.)
What about distance? That’s one of the things we were thinking of as a benefit – or a problem.
Say we’re slicing rutabagas and don’t feel like crossing the room to change from Pandora stations. Will it help from ten feet away or would we find ourselves shouting DELETE ROLLING INSERT SLY AND THE FAMILY Aahhhhrrrrrrrrr!
That was covered in my statement about noise ratio. The farther from the microphone and the more noise the worse the recognition. I couldn’t listen to anything like a television or stereo in the background unless I used a directional microphone directly in front of my mouth and had the set turned down low.
Get a trial version and use it. You will never be satisfied until you give it a try.