14thAnnual Conference of the International Speech Communication Association

Lyon, France
August 25-29, 2013

Musical Noise Analysis for Bayesian Minimum Mean-Square Error Speech Amplitude Estimators Based on Higher-Order Statistics

Hiroshi Saruwatari (1), Suzumi Kanehara (1), Ryoichi Miyazaki (1), Kiyohiro Shikano (1), Kazunobu Kondo (2)

(1) NAIST, Japan
(2) Yamaha, Japan

In this study, we perform a theoretical analysis of the amount of musical noise generated in Bayesian minimum mean-square error speech amplitude estimators. In our previous study, a musical noise assessment based on kurtosis has been successfully applied to spectral subtraction. However, it is difficult to apply this approach to the methods with a decision-directed a priori SNR estimator because it corresponds to a nonlinear recursive process for noise power spectral sequences. Therefore, in this paper, we analyze musical noise generation by combining Breithaupt-Martinfs approximation and our higher-order-statistics analysis. We also compare the result of theoretical analysis and that of objective experimental evaluation to indicate the validity of the proposed closed-form analysis.

Full Paper

Bibliographic reference.  Saruwatari, Hiroshi / Kanehara, Suzumi / Miyazaki, Ryoichi / Shikano, Kiyohiro / Kondo, Kazunobu (2013): "Musical noise analysis for Bayesian minimum mean-square error speech amplitude estimators based on higher-order statistics", In INTERSPEECH-2013, 441-445.