Tracking an Unknown Time-Varying Number of Speakers Using TDOA Measurements : A Random Finite Set Approach

W-K. Ma, Ba-Ngu Vo, S.S. Singh, Adrian Baddeley

    Research output: Contribution to journalArticle

    142 Citations (Scopus)

    Abstract

    Speaker location estimation techniques based on time-difference-of-arrival measurements have attracted much attention recently. Many existing localization ideas assume that only one speaker is active at a time. In this paper, we focus on a more realistic assumption that the number of active speakers is unknown and time-varying. Such an assumption results in a more complex localization problem, and we employ the random finite set (RFS) theory to deal with that problem. The RFS concepts provide us with an effective, solid foundation where the multispeaker locations and the number of speakers are integrated to form a single set-valued variable. By applying a sequential Monte Carlo implementation, we develop a Bayesian RFS filter that simultaneously tracks the time-varying speaker locations and number of speakers. The tracking capability of the proposed filter is demonstrated in simulated reverberant environments.
    Original languageEnglish
    Pages (from-to)3291-3304
    JournalIEEE Transactions on Signal Processing
    Volume54
    Issue number9
    DOIs
    Publication statusPublished - 2006

    Fingerprint

    Set theory
    Time difference of arrival

    Cite this

    Ma, W-K. ; Vo, Ba-Ngu ; Singh, S.S. ; Baddeley, Adrian. / Tracking an Unknown Time-Varying Number of Speakers Using TDOA Measurements : A Random Finite Set Approach. In: IEEE Transactions on Signal Processing. 2006 ; Vol. 54, No. 9. pp. 3291-3304.
    @article{2085b631d4f247a0864f469e48d81ae8,
    title = "Tracking an Unknown Time-Varying Number of Speakers Using TDOA Measurements : A Random Finite Set Approach",
    abstract = "Speaker location estimation techniques based on time-difference-of-arrival measurements have attracted much attention recently. Many existing localization ideas assume that only one speaker is active at a time. In this paper, we focus on a more realistic assumption that the number of active speakers is unknown and time-varying. Such an assumption results in a more complex localization problem, and we employ the random finite set (RFS) theory to deal with that problem. The RFS concepts provide us with an effective, solid foundation where the multispeaker locations and the number of speakers are integrated to form a single set-valued variable. By applying a sequential Monte Carlo implementation, we develop a Bayesian RFS filter that simultaneously tracks the time-varying speaker locations and number of speakers. The tracking capability of the proposed filter is demonstrated in simulated reverberant environments.",
    author = "W-K. Ma and Ba-Ngu Vo and S.S. Singh and Adrian Baddeley",
    year = "2006",
    doi = "10.1109/TSP.2006.877658",
    language = "English",
    volume = "54",
    pages = "3291--3304",
    journal = "IEEE Transactions on Signal Processing",
    issn = "1053-587X",
    publisher = "Wiley-IEEE Press",
    number = "9",

    }

    Tracking an Unknown Time-Varying Number of Speakers Using TDOA Measurements : A Random Finite Set Approach. / Ma, W-K.; Vo, Ba-Ngu; Singh, S.S.; Baddeley, Adrian.

    In: IEEE Transactions on Signal Processing, Vol. 54, No. 9, 2006, p. 3291-3304.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Tracking an Unknown Time-Varying Number of Speakers Using TDOA Measurements : A Random Finite Set Approach

    AU - Ma, W-K.

    AU - Vo, Ba-Ngu

    AU - Singh, S.S.

    AU - Baddeley, Adrian

    PY - 2006

    Y1 - 2006

    N2 - Speaker location estimation techniques based on time-difference-of-arrival measurements have attracted much attention recently. Many existing localization ideas assume that only one speaker is active at a time. In this paper, we focus on a more realistic assumption that the number of active speakers is unknown and time-varying. Such an assumption results in a more complex localization problem, and we employ the random finite set (RFS) theory to deal with that problem. The RFS concepts provide us with an effective, solid foundation where the multispeaker locations and the number of speakers are integrated to form a single set-valued variable. By applying a sequential Monte Carlo implementation, we develop a Bayesian RFS filter that simultaneously tracks the time-varying speaker locations and number of speakers. The tracking capability of the proposed filter is demonstrated in simulated reverberant environments.

    AB - Speaker location estimation techniques based on time-difference-of-arrival measurements have attracted much attention recently. Many existing localization ideas assume that only one speaker is active at a time. In this paper, we focus on a more realistic assumption that the number of active speakers is unknown and time-varying. Such an assumption results in a more complex localization problem, and we employ the random finite set (RFS) theory to deal with that problem. The RFS concepts provide us with an effective, solid foundation where the multispeaker locations and the number of speakers are integrated to form a single set-valued variable. By applying a sequential Monte Carlo implementation, we develop a Bayesian RFS filter that simultaneously tracks the time-varying speaker locations and number of speakers. The tracking capability of the proposed filter is demonstrated in simulated reverberant environments.

    U2 - 10.1109/TSP.2006.877658

    DO - 10.1109/TSP.2006.877658

    M3 - Article

    VL - 54

    SP - 3291

    EP - 3304

    JO - IEEE Transactions on Signal Processing

    JF - IEEE Transactions on Signal Processing

    SN - 1053-587X

    IS - 9

    ER -