Dereverberation by Using Time-Variant Nature of Speech Production System

• Published in: EURASIP journal on advances in signal processing Vol. 2007; no. 1
• Main Authors: Yoshioka, Takuya, Hikichi, Takafumi, Miyoshi, Masato
• Format: Journal Article
• Language: English
• Published: SpringerOpen 01.01.2007
• ISSN: 1687-6180; 1687-6172; 1687-6180
• DOI: 10.1155/2007/65698

This paper addresses the problem of blind speech dereverberation by inverse filtering of a room acoustic system.Since a speech signal can be modeled as being generated by a speech production system driven by an innovationsprocess, a reverberant signal is the output of a composite system consisting of the speech production and roomacoustic systems. Therefore, we need to extract only the part corresponding to the room acoustic system (or itsinverse filter) from the composite system (or its inverse filter). The time-variant nature of the speech productionsystem can be exploited for this purpose. In order to realize the time-variance-based inverse filter estimation,we introduce a joint estimation of the inverse filters of both the time-invariant room acoustic and the time-variantspeech production systems, and present two estimation algorithms with distinct properties.