First of all, thank you Chris for your very informative response. You are one of those who can help me with my specific questions.
Ok, i checked your version with the LixVoice, it’s more promissing then this one (which is by the way absolutely NOT USERFRIENDLY)…
But if i am honest, i have already generated something what has been working, until i changed something, what changed everything (and the bad thing is, i do not know anymore what i have changeg, so i can’t use the HTML-Component anymore at moment) to do the same thing, like shown in your EXAMPLE, but using some custom generated speech2text environment (html-component) and → elevenlabs.
So in detail, my setup is:
HTML-Component with some custom code inside, which listen to my mic.
It was working of somekind like…
(sorry for the bad voice-equalizer on screen → was just a test) 
Now you can see the spoken text in realtime inside of my HTML-Component.
Somehow i was able to activate that function and i got my speech to text inside the html in realtime, that means you were able to see each character like typed by a machine, but in real controlled by speech.
So, till here everything was working → WAS 
So i know, that it is possible to access through HTML-Component.
I just did not realize what i have done or what i have changed, so it was working.
So my questions at this point, regarding my setup…
a) Was the HTML-Component designed to allow microphone, camera, vr, speaker?
Because such thinngs are not included inside of the HTML-Component-API.
b) What would change, if i would use a CUSTOM-ELEMENT instead? Would it be easier for me to get those functions?
c) I like to use the HTML-Comp. because i am very familiar with this element right now, but i would switch over to CUSTOM-ELEMENT if this element provides more functionalty with less restrictions.
I also have seen some workarounds on how to give my own CUSTOM-ELEMENT the needed → allowens <— (inside of the origin-page-code) which did not really work well in most of cases.
The whole concept is connected to the following post…
d) How much → realtime ← is your version ?
e) what exactly would be possible if i would switch over to use your proposed version with lix-voice?
f) Would i be able for example just open a dropdown by my voice on my screen (this is what i am capable to do with my version)
g) Would i be able to create a new element on my screen, again just by my voice, i don’t think so. Yes, of course everything is running inside of my HTML-Element i know, but this is not a big problem, because at the end for real results i still can postMessage to get what i want on my Wix website.
h) And one more question about your provided version. Which languages does it support? (multi-lingual, like mine) ?
So, hopefully you recognize that i would have boundaries, if i would go your way.
It gives me some possibilities, but do not offer me the full package i need.
So for me it is important to be able to get that working on an HTML-Comp or custom element.
A third option maybe would be → NPM-package, but here again → NPM would directly run on my wix-page → again i wouldn’t have all the mentioned possibilities to get creative, since for example you CAN NOT create a new element by code, like .appendChild() → you know what i mean.
You also would not be able to reposition an element by speech on your site (maybe only if you would implement wix-animations somehow) .
And surely i could mention mcuh more situations, where your presented solution (good one), would not work for me (not flexible enough).
Or do you have other thoughts? If so → please let me know.
Best regards Dima