A method, computer program product, and system provide real time, two way audio-video conversation between mobile computing devices. Low latency is achieved by splitting the audio and video data streams from a given audio-video conversation using two different transport protocols to send the separate streams over a network, and re-syncing them at the other end. The transmission for each stream is tuned based on feedback data indicating available bandwidth of the network or other mobile computing device. A server offloads processing requirements that would otherwise be handled by the mobile computing device. The two way conversation can be externally observed by web-based users. The system functions over a disparate set of mobile computing device endpoints and web-based endpoints, and over different wireless carrier network infrastructures.