I am working on a demo site which includes a slide-out widget that allows a user to place a call.
I am using the SIPml5 tool along with the webrtc2sip back end for h
Well it is possible, but currently only in Chrome and with some assumptions. I am not the auther, you can find inspiration for this code in this open-source library (SimpleWebRtc).
navigator.webkitGetUserMedia(constraints,
function(webRTCStream){
var context = new window.AudioContext();
var microphone = context.createMediaStreamSource(webRTCStream);
var gainFilter = context.createGain();
var destination = context.createMediaStreamDestination();
var outputStream = destination.stream;
microphone.connect(gainFilter);
gainFilter.connect(destination);
var filteredTrack = outputStream.getAudioTracks()[0];
webRTCStream.addTrack(filteredTrack);
var originalTrack = webRTCStream.getAudioTracks()[0];
webRTCStream.removeTrack(originalTrack);
},
function(err) {
console.log("The following error occured: " + err);
}
);
The trick is to modify the stream and then replace the audio track of current stream with audio track of modified stream (taken from MediaStreamDestination stream).
DISCLAIMER:
This doesn't work in FireFox as of version 35, since they merely didn't implement MediaStream.addTrack/removeTrack. I use this check currently
this.micVolumeIsSupported = function() {
var MediaStream = window.webkitMediaStream || window.MediaStream;
return !!MediaStream.prototype.addTrack && !!MediaStream.prototype.removeTrack;
};
_gainSupported = this.micVolumeIsSupported();
This has a limitation in Chrome due to a bug with stopping stream with mixed up tracks. You might wish to restore these tracks before closing connection or on connection interruption;
this.restoreTracks = function(){
if(_gainSupported && _tracksSubstituted){
webRTCStream.addTrack(originalTrack);
webRTCStream.removeTrack(filteredTrack);
_tracksSubstituted = false;
}
};
This works for me