Windows 10 Voice recognition.

[BML]

I just tried the Windows 10 Voice recognition program out and I'm not impressed with the result. I have a small amount of Voice recognition experience from when I tried out what I think was called Dragon and I do understand that a Voice recognition program does require training. I gave up on Dragon because I simply did not have time to train it but its a good few years ago.
I have a number of diaries that I need to digitalise so that when I'm working on the content I have a search capability and I think that using a Voice recognition program would meet my needs but I have some questions.
Could anyone state whether in their experience the Windows 10 VR program could meet my needs?
If not do they know of a VR program that would?
Many thanks.

 

[Wolfie]

Hi BML,

is it set up correctly?:

https://support.microsoft.com/en-us/help/4027176/windows-10-use-voice-recognition

I don't use it btw so I have no idea how it functions at all!.

 

[BML]

With the greatest respect was there really any point in your reply?

 

[Trouble]

It would seem that you are looking for someone with current experience and like you, my personal experience is limited and somewhat dated, I started way back when Dragon Naturally Speaking was its' own company before Nuance bought them.
I do still from time to time test, play with and generally fiddle with current offerings.

Probably most critical is a good headset with built in microphone so both "training" and use remain consistent.
Such a device will also help dampen ambient noise which is also an important consideration.

I've found the native speech to text built into Windows 10 reasonably responsive as well as a good means of controlling your computer but maybe less so when doing dictation....

I have a number of diaries that I need to digitalise

IF your intention is to dictate these "diaries" into a digital document, the application you use to "dictate" is probably a very important consideration.
The current Microsoft 365 (Word and other application within the Suite) offering is a pretty good solution, but....
A product such as Dragon will likely be more robust as that's what they do.
They provide solutions for Medical as well as Legal environments where digital transcription is critically important.

Any solution or combination of solutions you choose is going to have a learning curve and some effort on your part both during the training as well as the actual use of the program.

Could anyone state whether in their experience the Windows 10 VR program could meet my needs?
If not do they know of a VR program that would?

My suggestion would be for you to do some experimenting on your own and determine what might be satisfactory (or not) based on your needs and how much time and energy you're willing to expend to accomplish the desired results.

Text to Speech products that are integrated into the Operating System or even the "dictation" feature in the newest MS Office product will probably never be as feature rich as a Software Product that is dedicated to such tasks.

Dictate text using Speech Recognition - Microsoft Support

Talk instead of typing your documents with Dictate or speech recognition.

support.office.com

 

[BML]

Probably most critical is a good headset with built in microphone so both "training" and use remain consistent.
Such a device will also help dampen ambient noise which is also an important consideration.

I've found the native speech to text built into Windows 10 reasonably responsive as well as a good means of controlling your computer but maybe less so when doing dictation....
IF your intention is to dictate these "diaries" into a digital document, the application you use to "dictate" is probably a very important consideration.
The current Microsoft 365 (Word and other application within the Suite) offering is a pretty good solution, but....
A product such as Dragon will likely be more robust as that's what they do.
They provide solutions for Medical as well as Legal environments where digital transcription is critically important.

Any solution or combination of solutions you choose is going to have a learning curve and some effort on your part both during the training as well as the actual use of the program.
My suggestion would be for you to do some experimenting on your own and determine what might be satisfactory (or not) based on your needs and how much time and energy you're willing to expend to accomplish the desired results.

Text to Speech products that are integrated into the Operating System or even the "dictation" feature in the newest MS Office product will probably never be as feature rich as a Software Product that is dedicated to such tasks.

Many thanks for your help. I will start my reply by updating where I got with the Windows 10 Speech Recognition which was not very far in that I didn’t even complete the tuning the microphone because the system kept packing in and as I couldn’t find a Windows 10 Speech Recognition forum I decided to dump it.

I was then told that I needed a USB Soundcard and I’ve ordered one.

You said, “Probably most critical is a good headset with built in microphone”

Why do I need a headset?

You said, “ the application you use to "dictate" is probably a very important consideration.”

I need to be precise in my understanding so could you please say what you mean by, “application”?

You appear to favour Dragon but would you care to say just which application do you suggest?
Many thanks for your help.
Regards, BML

 

[Trouble]

I was then told that I needed a USB Soundcard and I’ve ordered one.

No idea why that would be important unless your existing audio chipset was faulty?

Why do I need a headset?

It provides for consistency. You'll find that mic positioning is important and as opposed to a desktop mic or other integrated microphone it is easier to reposition the microphone in the same spot over and over again, and...
depending on its' design, may have other desirable properties such as inline volume and tonal controls as well as noise canceling ability to eliminate extraneous noises.

I need to be precise in my understanding so could you please say what you mean by, “application”?

I make that small distinction to hopefully clarify that a program (application) that is intended to accept dictation (voice (speech) input) and use that to generate a text document as in TEXT to SPEECH
IS
Slightly different than another piece of software that may be more targeted toward computer functions and operations (as in an ease of access kind of way) and may be slightly less adequate for creating an actual text document.

I've notice people complaining about the Windows 10 VR software glitching out on some document commands involving punctuation(s) or other commands like "new line", etc.

You appear to favour Dragon

Nope, I really don't and I also don't have any recommendations.
I find most such products a bit cumbersome, although I wouldn't hesitate to employ one of them if for some reason, I needed it.
BUT
I'm so lazy, I would likely just use a digital recorder and digitize them as audio files
AND
IF I really needed an actual text document, I'd look for something I could just drop that audio file into and it would produce a text document from that. I got to believe that there is some such product out there somewhere

 

[BML]

No idea why that would be important unless your existing audio chipset was faulty?

It provides for consistency. You'll find that mic positioning is important and as opposed to a desktop mic or other integrated microphone it is easier to reposition the microphone in the same spot over and over again, and...
depending on its' design, may have other desirable properties such as inline volume and tonal controls as well as noise canceling ability to eliminate extraneous noises.

I make that small distinction to hopefully clarify that a program (application) that is intended to accept dictation (voice (speech) input) and use that to generate a text document as in TEXT to SPEECH
IS
Slightly different than another piece of software that may be more targeted toward computer functions and operations (as in an ease of access kind of way) and may be slightly less adequate for creating an actual text document.

I've notice people complaining about the Windows 10 VR software glitching out on some document commands involving punctuation(s) or other commands like "new line", etc.

Nope, I really don't and I also don't have any recommendations.
I find most such products a bit cumbersome, although I wouldn't hesitate to employ one of them if for some reason, I needed it.
BUT
I'm so lazy, I would likely just use a digital recorder and digitize them as audio files
AND
IF I really needed an actual text document, I'd look for something I could just drop that audio file into and it would produce a text document from that. I got to believe that there is some such product out there somewhere

This where I am so far.

I attempted to set up the microphone as its part of the Windows 10 Speech Recognition process but only got halfway into the procedure when it stopped so I asked the following question on Forum that I use.

“My mic has the following words round the top. ALTIDMV - 100 Uni-directional IMP - 600 omp. Can anyone tell me if that will work on Windows 10 Speech Recognition.” I received this reply. “This microphone should work if you additionally purchase a 3.5 mm adapter plug and a USB Soundcard. Note that there is a difference between “should work” and “ideally suited”.”

I then said, “I have a supplementary question. You said. "Note that there is a difference between "should work" and "ideally suited." What would be an ideally suited solution?” I received this answer, “We probably could've made our answer clearer if we said OK vs ideal. With the right attachments, your current microphone should work fine but not quite as good as a proper USB Microphone that is designed with speech recognition in mind so it looks as though I should get one of those.

I also asked the following question. “Why do I need a headset?” I received this answer. “It provides for consistency.” “You'll find that mic positioning is important and as opposed to a desktop mic or other integrated microphone it is easier to reposition the microphone in the same spot over and over again, and...
depending on its' design, may have other desirable properties such as inline volume and tonal controls as well as noise cancelling ability to eliminate extraneous noises.”

I then asked, “I need to be precise in my understanding so could you please say what you mean by, “application”?

I received this answer. “I make that small distinction to hopefully clarify that a program (application) that is intended to accept dictation (voice (speech) input) and use that to generate a text document as in TEXT to SPEECH IS Slightly different that another piece of software that may be more targeted toward computer functions and operations and may be slightly less adequate for creating an actual text document.”

Yesterday I got stuck just attempting to set up the microphone but today I simply switched the Windows Speech icon on to “On: Listen to everything I say.” And the first line that I input showed without a fault. As I have ordered a 3.5 mm adapter plug and a USB Soundcard I will give it a try and I think that its worth trying a “proper USB Microphone that is designed with speech recognition in mind”. The only problem being from my point of view is how do I ensure that I actually get a, “proper USB Microphone that is designed with speech recognition in mind”.

Thank you very much for the help you have given me so far. BML

 

[Trouble]

how do I ensure that I actually get a, “proper USB Microphone that is designed with speech recognition in mind”.

I don't know.
I have a USB Microphone, it's handheld, I bought it for my grand daughter so her and her sixth grade posse could fool around with Karaoke. Personally, I neve use it.
My guess is that whoever supplied that suggestion/answer is likely aware that there are certainly some microphones available that are better, higher quality and more accurate in voice reproduction, due to their construction and components used.
They are likely more expensive as well.

I wouldn't consider investing big bucks in equipment until I performed some research and experimentation with what I had on hand and determined if my needs were such that something of better quality was required to justify the added expense.

I can certainly appreciate the value of quality equipment, but....
You probably don't actually need sound studio level equipment. You're not attempting to produce an audio book.
As I've mentioned before, experiment on your own and see what the end product provides.
Some stumbling and bumbling can be expected and ultimately provides the education and skills you'll need going forward.

 

[BML]

A final question I hope. Would one get a better effect by using a Windows 10 dedicated USB microphone?