Theoretical analysis of parametric blind spatial subtraction array and its application to speech recognition performance prediction

Author

Miyazaki, Ryoichi ; Saruwatari, Hiroshi ; Wakisaka, Ryo ; Shikano, Kiyohiro ; Takatani, Tomoya

Author_Institution

Nara Inst. of Sci. & Technol., Nara, Japan

fYear

2011

fDate

May 30 2011-June 1 2011

Firstpage

19

Lastpage

24

Abstract

In this paper, an improved parametric postfiltering is introduced in our previously proposed blind spatial subtraction array (BSSA), and its theoretical analysis of the amounts of musical noise and noise reduction is conducted via higher-order statistics. Compared with the conventional BSSA, it is clarified that parametric BSSA can improve speech recognition performance. Next, we propose an unsupervised speech-recognition-performance prediction metric based on higher-order statistics in BSSA. We successfully reveal that the noise and speech kurtosis can be used for predicting speech recognition performance without using any reference speech signals.

Keywords

Wiener filters; filtering theory; higher order statistics; signal denoising; speech recognition; higher-order statistics; musical noise; noise reduction; parametric blind spatial subtraction array; parametric postfiltering; quasi-parametric Wiener filter; reference speech signals; speech kurtosis; unsupervised speech-recognition-performance prediction metric; Accuracy; Gaussian noise; Noise reduction; Speech; Speech enhancement; Speech recognition; Higher-order statistics; Musical noise; Speech enhancement; Speech recognition performance prediction;

fLanguage

English

Publisher

ieee

Conference_Titel

Hands-free Speech Communication and Microphone Arrays (HSCMA), 2011 Joint Workshop on

Conference_Location

Edinburgh

Print_ISBN

978-1-4577-0997-5

Type

conf

DOI

10.1109/HSCMA.2011.5942397

Filename

5942397