A Reconfigurable Codec for Secure Voice Communication

A novel voice communication scheme is presented which assures security in voice communication well beyond any encryption-based method in use today. The scheme is based on an original human voice production model called Sympes, which posits that human speech can be decomposed into and reconstructed from a set of so-called energy and information functions. It was discovered that a finite set of these functions is sufficient to represent any and all human speech regardless of language, gender, age, or ethnic background. The model renders itself to be used as a voice codec which can be configured in an almost infinite number of ways, many orders of magnitude larger than the key space afforded by any current encryption algorithm. This paper explains the Sympes model, how it was used to develop a co dec, and how that codec formed the basis of a secure voice communication app called Symphone, working in real time on Android, iOS, and Windows platforms. Symphone is currently in use in certain undisclosed security applications and is available to institutions worldwide.