Hi everyone,
how do we detect user voice in real time. When ever user speaking until user hangup the call I need to detect user speaking or not.
This is now becoming at a minimum a weekly question at this point alongside AI. Asterisk provides the TALK_DETECT dialplan function[1]. If this is insufficient then either you have to extend its capabilities, or do such a thing outside of Asterisk.
I am created real time voice to voice communication custom application through ARI and sockets streaming but i don`t know how do i detect user voice in real time.The requirement is when ever user speaking until hangup the call. I need to detect the voice accordingly
From a theoretical point of view, you can’t detect a voice in real time, as you can only tell it is a voice, and not another source of noise, some time after the start. Even just detecting sound level cannot be done instantaneously.
So what do you mean by voice here, and what do you mean by real time? (This is not to say that there will be an easy Asterisk solution, even given clarification.)
Also, I’m having difficulty understand what value Asterisk adds to your application, as it sounds like you are bypassing most of the things it does well.