currently i am testing all user media elements.
Ive managed to stream pictures to a server and display them on other devices, pretty straigth forward. i tryed the same with audio, but as i see it there is no way to do it with the current tools. so i thought at least the microphone would be usefull to automatically start speech recognition.
so i measure the rms and when rms>-20 i would like to start the recognition.
but what i experienced was that the browser asks EVERY TIME the recvognition starts for permission. is there a way to only once ask for permission and then just start / stop recognition? that would be an important feature. it makes litterally no sence to request it for every sentence