Speech Enhancement Using ES Information and Phase-Error Based Filters

This paper is concerned with speech enhancement using Phase-Error based Filters (PEF) and Excitation Source (ES) information in car environments. For this purpose, we firstly use ES information to determine the time-delay from speech signals obtained by two microphones for sound source localization. Then, the phase-error based filters are performed by prior knowledge regarding the time delay obtained by ES information and the phases of the signals recorded by the microphones. The experimental results showed the effectiveness of the presented method for speech enhancement.

[1]  Yoonho Jang,et al.  The development of the vehicle sound source localization system , 2015, 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA).

[2]  Hanseok Ko,et al.  Sudden noise source localization system for intelligent automobile application with acoustic sensors , 2012, 2012 IEEE International Conference on Consumer Electronics (ICCE).

[3]  Sung-Suk Kim,et al.  Sound source localization with the aid of excitation source information in home robot environments , 2008, IEEE Transactions on Consumer Electronics.

[4]  S. R. Mahadeva Prasanna,et al.  Speaker localization using excitation source information in speech , 2005, IEEE Transactions on Speech and Audio Processing.

[5]  Bayya Yegnanarayana,et al.  Determining Number of Speakers From Multispeaker Speech Signals Using Excitation Source Information , 2007, IEEE Signal Processing Letters.

[6]  Minsoo Hahn,et al.  Dual-Microphone Noise Reduction in Car Environments With Determinant Analysis of Input Correlation Matrix , 2016, IEEE Sensors Journal.

[7]  Parham Aarabi,et al.  Phase-based dual-microphone robust speech enhancement , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).