Abstract: In this work, speech signals are modeled by means of so called the pre-defined "signature functions". The pre-defined signature functions are generated using the statistical properties of the speech signals. It has been exhibited that, with a few basic signature functions, any hearable sounds such as music and voice can be generated within a tolerable error.
Rafet Akdeniz and B. Siddik Yarman , 2004. A New Statistical Method to Represent Speech Signals . Asian Journal of Information Technology, 3: 721-726.