Two versions of Windows speech recognition?

RadxBoy · 02 Oct 2020

There appears to be at least two versions of speech recognition in my version of Windows 10 (Version: 10.0.18362 Build 18362). One is an app which looks like this:

Two versions of Windows speech recognition?-speech-app1.png

and the other is invoked with Win + H and looks like this:

Two versions of Windows speech recognition?-speech-app2-using-win-h.png

.

Does anyone know the difference between these and which one we should be using? I don't want to get used to one and then have MS remove it in an update!

Note: They appear to be different because I had to "train" the first before it would work well; the other (Win + H) did not appear to need any training!

Thanks!

dalchina · 02 Oct 2020

The first dates from Vista and has a fair bit of support- standard phrases on a reference card which can be used to control Windows dialogues e.g.

Two versions of Windows speech recognition?-1.png

Tutorials in the Tutorial section:

The second is, I think a UWP app and started with Win 10. It can't be used to control Windows. It is less sophisticated.
How to Use Dictation in Windows 10

RadxBoy · 07 Oct 2020

One thing I found out, the first app is actually a program within Windows which performs speech recognition, the second (Win + H) is an online application which sends your voice to Microsoft. If you have privacy concerns, this might not be the way to go.

dalchina · 07 Oct 2020

an online application which sends your voice to Microsoft

- which is equally true of Cortana... if you're concerned about such things I suppose you might start thinking Cortana starts with a 'c' - and so does conspiracy...