Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones

Mitsuharu Matsumoto, Shuji Hashimoto

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    This paper introduces a blind source separation (BSS) algorithm in the time domain based on the amplitude gain difference of two directional microphones located at the same place, namely aggregated microphones. A feature of our approach is to treat the BSS problem of the anechoic mixtures in the time domain. Sparseness approach is one of the attractive methods to solve the problem of the sound separation. If the signal is sparse in the frequency domain, the sources rarely overlap. Under this condition, it is possible to extract each signal using time-frequency binary masks. In this paper, we treat the non-stationary, partially disjoint signals. In other words, most of the signals overlap in the time domain and the frequency domain though there exist some intervals where the sound is disjoint. We firstly show the source separation problem can be described not as a convolutive model but as an instantaneous model in spite of the anechoic mixing when the aggregated microphones are assumed. We then show the necessary conditions and show the algorithm with the experimental results. In this method, we can treat the problem not in the time-frequency domain but in the time domain due to the characteristics of the aggregated microphones. In other words, we can consider the problem not in the complex space but in the real space. The mixing matrix can be directly identified utilizing the observed signals without estimating the intervals where the signal is disjoint through all the processes.

    Original languageEnglish
    Title of host publicationEuropean Signal Processing Conference
    Publication statusPublished - 2006
    Event14th European Signal Processing Conference, EUSIPCO 2006 - Florence, Italy
    Duration: 2006 Sep 42006 Sep 8

    Other

    Other14th European Signal Processing Conference, EUSIPCO 2006
    CountryItaly
    CityFlorence
    Period06/9/406/9/8

    Fingerprint

    Blind source separation
    Microphones
    Acoustic waves
    Source separation
    Masks

    ASJC Scopus subject areas

    • Signal Processing
    • Electrical and Electronic Engineering

    Cite this

    Matsumoto, M., & Hashimoto, S. (2006). Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones. In European Signal Processing Conference

    Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones. / Matsumoto, Mitsuharu; Hashimoto, Shuji.

    European Signal Processing Conference. 2006.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Matsumoto, M & Hashimoto, S 2006, Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones. in European Signal Processing Conference. 14th European Signal Processing Conference, EUSIPCO 2006, Florence, Italy, 06/9/4.
    Matsumoto M, Hashimoto S. Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones. In European Signal Processing Conference. 2006
    Matsumoto, Mitsuharu ; Hashimoto, Shuji. / Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones. European Signal Processing Conference. 2006.
    @inproceedings{7cb7bd085a6e4454b58d9a790870f087,
    title = "Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones",
    abstract = "This paper introduces a blind source separation (BSS) algorithm in the time domain based on the amplitude gain difference of two directional microphones located at the same place, namely aggregated microphones. A feature of our approach is to treat the BSS problem of the anechoic mixtures in the time domain. Sparseness approach is one of the attractive methods to solve the problem of the sound separation. If the signal is sparse in the frequency domain, the sources rarely overlap. Under this condition, it is possible to extract each signal using time-frequency binary masks. In this paper, we treat the non-stationary, partially disjoint signals. In other words, most of the signals overlap in the time domain and the frequency domain though there exist some intervals where the sound is disjoint. We firstly show the source separation problem can be described not as a convolutive model but as an instantaneous model in spite of the anechoic mixing when the aggregated microphones are assumed. We then show the necessary conditions and show the algorithm with the experimental results. In this method, we can treat the problem not in the time-frequency domain but in the time domain due to the characteristics of the aggregated microphones. In other words, we can consider the problem not in the complex space but in the real space. The mixing matrix can be directly identified utilizing the observed signals without estimating the intervals where the signal is disjoint through all the processes.",
    author = "Mitsuharu Matsumoto and Shuji Hashimoto",
    year = "2006",
    language = "English",
    booktitle = "European Signal Processing Conference",

    }

    TY - GEN

    T1 - Blind source separation of anechoic mixtures in time domain utilizing aggregated microphones

    AU - Matsumoto, Mitsuharu

    AU - Hashimoto, Shuji

    PY - 2006

    Y1 - 2006

    N2 - This paper introduces a blind source separation (BSS) algorithm in the time domain based on the amplitude gain difference of two directional microphones located at the same place, namely aggregated microphones. A feature of our approach is to treat the BSS problem of the anechoic mixtures in the time domain. Sparseness approach is one of the attractive methods to solve the problem of the sound separation. If the signal is sparse in the frequency domain, the sources rarely overlap. Under this condition, it is possible to extract each signal using time-frequency binary masks. In this paper, we treat the non-stationary, partially disjoint signals. In other words, most of the signals overlap in the time domain and the frequency domain though there exist some intervals where the sound is disjoint. We firstly show the source separation problem can be described not as a convolutive model but as an instantaneous model in spite of the anechoic mixing when the aggregated microphones are assumed. We then show the necessary conditions and show the algorithm with the experimental results. In this method, we can treat the problem not in the time-frequency domain but in the time domain due to the characteristics of the aggregated microphones. In other words, we can consider the problem not in the complex space but in the real space. The mixing matrix can be directly identified utilizing the observed signals without estimating the intervals where the signal is disjoint through all the processes.

    AB - This paper introduces a blind source separation (BSS) algorithm in the time domain based on the amplitude gain difference of two directional microphones located at the same place, namely aggregated microphones. A feature of our approach is to treat the BSS problem of the anechoic mixtures in the time domain. Sparseness approach is one of the attractive methods to solve the problem of the sound separation. If the signal is sparse in the frequency domain, the sources rarely overlap. Under this condition, it is possible to extract each signal using time-frequency binary masks. In this paper, we treat the non-stationary, partially disjoint signals. In other words, most of the signals overlap in the time domain and the frequency domain though there exist some intervals where the sound is disjoint. We firstly show the source separation problem can be described not as a convolutive model but as an instantaneous model in spite of the anechoic mixing when the aggregated microphones are assumed. We then show the necessary conditions and show the algorithm with the experimental results. In this method, we can treat the problem not in the time-frequency domain but in the time domain due to the characteristics of the aggregated microphones. In other words, we can consider the problem not in the complex space but in the real space. The mixing matrix can be directly identified utilizing the observed signals without estimating the intervals where the signal is disjoint through all the processes.

    UR - http://www.scopus.com/inward/record.url?scp=84862611386&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84862611386&partnerID=8YFLogxK

    M3 - Conference contribution

    BT - European Signal Processing Conference

    ER -