I've used the MediaStreamTrack API to get the list of audio and video capture devices to allow users to control the input device for WebRTC calls. See
http://stackoverflow.com/a/20087996/552792 for an example.
However, this doesn't always seem to affect the output device. We change the "Default" input device to, let's say, "Logitech USB Headset", and the input changes from the default to the headset, but the output still comes out of the speaker.
I don't think WebRTC (getUserMedia and PeerConnection) interoperate with the output device, but I assume there must be some way to control this from Node.js, node-webkit, or a Chrome API. Is this possible? If so, how should I get started?
Thank you!
James Mortensen