Is it possible to disable the screen element reader and only keep the keyboard key press sound?

[pedoc]

thanks a lot

[SaschaCowley]

@pedoc could you please rephrase your query? I am not sure I understand what you're asking.

[pedoc]

Excuse me, let me rephrase my motivation.
Currently, I'm looking for a software that can play the pressed keys, for example, when I press A, the sound A will play.
Then, I found nvda, which meets the requirements, but I don't need to read the screen content (regardless of the mouse focus or the focus controlled by the keyboard). This is for children to learn letters, and to protect their eyesight, the screen will be turned off, but the focus switch will cause the screen content to be read (even if the screen is turned off), which brings some additional distractions.

So, I hope to only keep key capture and sound playback, and turn off the screen reading function.

I tried to write a small program using c# winform, but because the program priority was not high enough, it often lost focus and could not capture keystrokes. I guess nvda and other software should use special APIs.

thanks

[valiant8086]

you can probably do that with autoit. Check for keyDown and then loop until keyUp and announce the key. Something like that anyway as opposed to keyPressed. Speak the letters through SAPI5, or record audio clips and play those. Not sure if you'll get far wanting to have a screen reader not be a screen reader. Btw, there should be ways for your test application to regain focus, I think. Check if it's not maximized and just re-maximize it. But that would get pretty irritating if you actually wanted to go somewhere else lol.

[pedoc]

I am more familiar with C#. At the same time, AutoIt should not have a higher priority than the WinForm program.

I set TopMost=true in the program and displayed it in full screen, but you know, children sometimes trigger random keystrokes such as Win+D. The WinForm program cannot intercept this key combination, causing the program to lose focus.

Maybe a driver-level program is needed to intercept, but this is difficult to implement (EV signature is too expensive)

So, I am also trying other solutions, such as buying a Raspberry Pi and deploying the program on the Raspberry Pi, but so far, there has been no progress 🤣

[SaschaCowley]

I'm afraid I'm not aware of any way of achieving what you want. You could try using Input help, though this will report the functions of NVDA's gestures, as well as the names of the keys, and students will be able to exit input help with NVDA+1. It also doesn't announce the result of modified keys (that is, you'll hear "Shift plus 1", not "exclamation mark").

At the end of the day, I think this usage is really trying to use a hammer to screw in a screw. You're constantly going to be fighting what NVDA is designed to do.

[pedoc]

Thanks for the help, I'll try to solve it myself ❤️

[valiant8086]

Hi.

I did manage to make a small autoit3 application which provides keyboard echo with my voice, but if I type really really fast the characters get entered in the wrong order sometimes, so I'm not sure what to do about that. But it works pretty much everywhere, will read the letters you type into notepad, on the desktop, etc.

Only place it hasn't been reading so far is task manager. But in any case, even if you or I can't get that to work reliably, I guess this proves that it is possible to do global keyboard capturing for this purpose. AT the moment my code is a bit messed up because I changed how it works mid-way through so once I clean it up I can send a link for it along.

[seanbudd]

Locking as off-topic to NVDA