Title: Distant Sound Suppression Using Spectral Phase Variance for Two Channel Blind Source Separation
Abstract: Various techniques of sound source separation have been proposed. Generally, they may try to separate and extract all the sound sources included in observed signals. On some applications, e.g., speech recognition systems, it requires to extract only desired sound sources, and other sound sources are not important. Hence, it is useful a system that evaluates the importance of each separated sound source and extracts ones with high importance. In this paper, it is assumed that the importance of sound source becomes low according with its distance to the microphone. The proposed method estimates the distance of each sound source by using the spectral phase variance which is calculated with the phase difference between the two microphones. Based on the estimated distance, we suppress distant sound sources and extract ones existing within the desired region.
Publication Year: 2018
Publication Date: 2018-11-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
Cited By Count: 5
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot