Mid-session Audio Codec Change

The SteganRTP reference implementation's embedding component addresses the issue of mid-session audio Codec change by determining the audio sample word size dynamically based on the Codec value supplied by the RTP packet's header. Thus, the embedding system's parameters are derived from each individual RTP packet that will be embedded into as cover-medium. If the RTP session were to change Codecs mid-session, or even to change Codecs for every other packet, the embedding system will only operate on RTP packets who's payloads are encoded with a Codec that the embedding system recognizes and has parameters defined for. If the embedding system does not recognize and support a particular packet's Codec, that packet is passed unmodified.