Milos Cernak, Afsaneh Asaei, and Alexandre Hyafil Cognitive Speech Coding Examining the impact of cognitive speech processing on speech compression S peech coding is a field in which compression paradigms have not changed in the last 30 years. Speech signals are most commonly encoded with compression methods that have roots in linear predictive theory dating back to the early 1940s. This article bridges this influential theory with recent cognitive studies applicable in speech communication engineering. It reviews the mechanisms of speech perception that have led to perceptual speech coding. The focus is on human speech communication and machine learning and the application of cognitive speech processing in speech compression that presents a paradigm shift from perceptual (auditory) speech processing toward cognitive (auditory plus cortical) speech processing. Our aim is to provide an overview of the impact of cognitive speech processing on speech compression and to discuss challenges faced in this interdisciplinary speech processing field. In this context, the traditional speech coding techniques as well as emerging approaches facilitated by deep-learning computational methods are covered. Key references are identified on fundamental teachings of psycholinguistics and speech neuroscience, and a valuable background for beginners and practitioners is given regarding the promising directions of incorporating principles of cognitive speech processing in speech compression. Digital Object Identifier 10.1109/MSP.2017.2761895 Date of publication: 26 April 2018 HEAD: ©iStockpHoto.com/pitJU; SoUND WAVE: ©iStockpHoto.com/ANDREY SUSLoV 1053-5888/18©2018IEEE IEEE Signal Processing Magazine | May 2018 | 97http://www.iStockpHoto.com/pitJU http://www.iStockpHoto.com/ANDREY