Sound source separation using null-beamforming and spectral subtraction for mobile devices

Shintaro Takada, Satoshi Kanba, Tetsuji Ogawa, Kenzo Akagiri, Tetsunori Kobayashi

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

    7 Citations (Scopus)

    Abstract

    This paper presents a new type of speech segregation method for mobile devices in noisy sound situation, where two or more speakers are talking simultaneously. The proposed method consists of multiple null-beamformers, their minimum power channel selection and spectral subtraction. The proposed method is performed with space-saving and coplanar microphone arrangements and low-cost calculations, which are the very important requirements for the mobile application. Effectiveness of the proposed method is clarified in the segregation and the recognition experiments of two simultaneous continuous speeches: the method improved the PESQ-based MOS value by about one point and reduced 70% of word recognition errors compared with non-processing.

    Original languageEnglish
    Title of host publicationIEEE Workshop on Applications of Signal Processing to Audio and Acoustics
    Pages30-33
    Number of pages4
    DOIs
    Publication statusPublished - 2007
    Event2007 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA - New Paltz, NY
    Duration: 2007 Oct 212007 Oct 24

    Other

    Other2007 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA
    CityNew Paltz, NY
    Period07/10/2107/10/24

    Fingerprint

    Source separation
    Beamforming
    Mobile devices
    Acoustic waves
    Microphones
    Costs
    Experiments

    ASJC Scopus subject areas

    • Signal Processing

    Cite this

    Takada, S., Kanba, S., Ogawa, T., Akagiri, K., & Kobayashi, T. (2007). Sound source separation using null-beamforming and spectral subtraction for mobile devices. In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (pp. 30-33). [4393006] https://doi.org/10.1109/ASPAA.2007.4393006

    Sound source separation using null-beamforming and spectral subtraction for mobile devices. / Takada, Shintaro; Kanba, Satoshi; Ogawa, Tetsuji; Akagiri, Kenzo; Kobayashi, Tetsunori.

    IEEE Workshop on Applications of Signal Processing to Audio and Acoustics. 2007. p. 30-33 4393006.

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

    Takada, S, Kanba, S, Ogawa, T, Akagiri, K & Kobayashi, T 2007, Sound source separation using null-beamforming and spectral subtraction for mobile devices. in IEEE Workshop on Applications of Signal Processing to Audio and Acoustics., 4393006, pp. 30-33, 2007 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA, New Paltz, NY, 07/10/21. https://doi.org/10.1109/ASPAA.2007.4393006
    Takada S, Kanba S, Ogawa T, Akagiri K, Kobayashi T. Sound source separation using null-beamforming and spectral subtraction for mobile devices. In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics. 2007. p. 30-33. 4393006 https://doi.org/10.1109/ASPAA.2007.4393006
    Takada, Shintaro ; Kanba, Satoshi ; Ogawa, Tetsuji ; Akagiri, Kenzo ; Kobayashi, Tetsunori. / Sound source separation using null-beamforming and spectral subtraction for mobile devices. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics. 2007. pp. 30-33
    @inproceedings{0c99881259e34e16b16ae7bb7f33b5ea,
    title = "Sound source separation using null-beamforming and spectral subtraction for mobile devices",
    abstract = "This paper presents a new type of speech segregation method for mobile devices in noisy sound situation, where two or more speakers are talking simultaneously. The proposed method consists of multiple null-beamformers, their minimum power channel selection and spectral subtraction. The proposed method is performed with space-saving and coplanar microphone arrangements and low-cost calculations, which are the very important requirements for the mobile application. Effectiveness of the proposed method is clarified in the segregation and the recognition experiments of two simultaneous continuous speeches: the method improved the PESQ-based MOS value by about one point and reduced 70{\%} of word recognition errors compared with non-processing.",
    author = "Shintaro Takada and Satoshi Kanba and Tetsuji Ogawa and Kenzo Akagiri and Tetsunori Kobayashi",
    year = "2007",
    doi = "10.1109/ASPAA.2007.4393006",
    language = "English",
    isbn = "9781424416196",
    pages = "30--33",
    booktitle = "IEEE Workshop on Applications of Signal Processing to Audio and Acoustics",

    }

    TY - GEN

    T1 - Sound source separation using null-beamforming and spectral subtraction for mobile devices

    AU - Takada, Shintaro

    AU - Kanba, Satoshi

    AU - Ogawa, Tetsuji

    AU - Akagiri, Kenzo

    AU - Kobayashi, Tetsunori

    PY - 2007

    Y1 - 2007

    N2 - This paper presents a new type of speech segregation method for mobile devices in noisy sound situation, where two or more speakers are talking simultaneously. The proposed method consists of multiple null-beamformers, their minimum power channel selection and spectral subtraction. The proposed method is performed with space-saving and coplanar microphone arrangements and low-cost calculations, which are the very important requirements for the mobile application. Effectiveness of the proposed method is clarified in the segregation and the recognition experiments of two simultaneous continuous speeches: the method improved the PESQ-based MOS value by about one point and reduced 70% of word recognition errors compared with non-processing.

    AB - This paper presents a new type of speech segregation method for mobile devices in noisy sound situation, where two or more speakers are talking simultaneously. The proposed method consists of multiple null-beamformers, their minimum power channel selection and spectral subtraction. The proposed method is performed with space-saving and coplanar microphone arrangements and low-cost calculations, which are the very important requirements for the mobile application. Effectiveness of the proposed method is clarified in the segregation and the recognition experiments of two simultaneous continuous speeches: the method improved the PESQ-based MOS value by about one point and reduced 70% of word recognition errors compared with non-processing.

    UR - http://www.scopus.com/inward/record.url?scp=50249136074&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=50249136074&partnerID=8YFLogxK

    U2 - 10.1109/ASPAA.2007.4393006

    DO - 10.1109/ASPAA.2007.4393006

    M3 - Conference contribution

    AN - SCOPUS:50249136074

    SN - 9781424416196

    SP - 30

    EP - 33

    BT - IEEE Workshop on Applications of Signal Processing to Audio and Acoustics

    ER -