Abstract
Speech enhancement can be achieved in temporal, spectral, and spatial domains. For practical use, spectral subtraction is a popular filter in the spectral domain, and beamforming is a representative spatial filter. However, the former assumes stationary background noise, which should be known or estimated, and the latter requires arrival directions on sound sources. The author has proposed a robust approach to the simultaneous estimation of both a direction‐of‐arrival (DOA) of a target signal and an amplitude spectrum of background noise [Mizumachi, Proc. 158th ASA meeting]. The proposed method has also given confidence for each DOA estimate. In this paper, a spatial filter is adaptively designed and is integrated with a spectral filter in the adaptation process on spectral‐spatial space. The confidence for each DOA estimate plays an important role for designing a reasonable and efficient spectral‐spatial filter under time‐varying acoustic scenes. In other words, simultaneous adaptation can be achieved time by time based on the reliability of each DOA estimate for the 2‐D spectral‐spatial filter. [Work supported by NEDO, Japan.]
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
