I am considering the issues and options concerning a feature that we intend to add to a Mobile AIR app. The feature would involve enabling the app user to purchase a video or voice/avatar discussion with a remote "tutor".
I am aware of two limitations in Adobe AIR with regards to voice or video chat. So, I am thinking about work-arounds, and I'd like to compare notes with other developers, see if anyone else has any experience or ideas that might help - and check that I'm going in the right direction with this.
1. The first limitation is that Adobe AIR has echo cancellation for desktop apps, but not mobile apps. Assuming that the client is using a mobile app, and the "tutor" is using a desktop app. So echo cancellation would just be at the desktop/tutor end. Am I correct in assuming that the client will be less aware of the echo than the tutor?? (This is what I'd prefer, given that the client would be paying for the service, and the tutor could just put up with it.)
2. The second limitation is that hardware accelerated video decoding of RTMP streams is not possible on iOS. For this reason, I'm considering, just voice, or a simple avatar with mouth size driven by the (mean squared) power in a block of voice signal values. Has anyone else done this? I assume it is possible to drive this real-time, synchronised with the sound?
3. Finally, I might also want to drive some sort of "whiteboard" functionality. This entails some extra data sent from the "tutor" to the client mobile app. The data would be interpreted as whiteboard data and rendered on the client device. Again, I'd be interested in discussing with other developers who might have done something similar.