Reduction of audio truncation when transcoding speech data from one coding format to another. Embodiments include receiving packets of a first communication session containing first encrypted speech data encoded according to a vocoder of a first type and encrypted using an encryption protocol, and containing a first encryption protocol identification information distributed among the received packets. Further embodiments include extracting the first encryption protocol identification information from the received packets and processing the received packets based on the first encryption identification information. Embodiments include transmitting one or more voice header packets containing a second encryption protocol identification information in a second communication session as well as transmitting packets in the second communication session containing second encrypted speech data encoded according to a vocoder of a second type.