Confidence Based Acoustic Event Detection

    Research output: Chapter in Book/Conference paperConference paper

    4 Citations (Scopus)

    Abstract

    Acoustic event detection, the determination of the acoustic event type and the localisation of the event, has been widely applied in many real-world applications. Many works adopt the multi-label classification technique to perform the polyphonic acoustic event detection with a global threshold to detect the active acoustic events. However, the manually labeled boundaries are error-prone and cannot always be accurate, especially when the frame length is too short to be accurately labeled by human annotators. To deal with this, a confidence is assigned to each frame and acoustic event detection is performed using a multi-variable regression approach in this paper. Experimental results on the latest TUT sound event 2017 database of polyphonic events demonstrate the superior performance of the proposed approach compared to the multi-label classification based AED method.

    Original languageEnglish
    Title of host publication2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
    Place of PublicationUSA
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages306-310
    Number of pages5
    Volume2018-April
    ISBN (Print)9781538646588
    DOIs
    Publication statusPublished - 10 Sep 2018
    Event2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Calgary, Canada
    Duration: 15 Apr 201820 Apr 2018

    Conference

    Conference2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018
    CountryCanada
    CityCalgary
    Period15/04/1820/04/18

    Fingerprint

    Acoustics
    Labels
    Acoustic waves

    Cite this

    Xia, X., Togneri, R., Sohel, F., & Huang, D. (2018). Confidence Based Acoustic Event Detection. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings (Vol. 2018-April, pp. 306-310). [8461845] USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICASSP.2018.8461845
    Xia, Xianjun ; Togneri, Roberto ; Sohel, Ferdous ; Huang, David. / Confidence Based Acoustic Event Detection. 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Vol. 2018-April USA : IEEE, Institute of Electrical and Electronics Engineers, 2018. pp. 306-310
    @inproceedings{926bd6a79f18455c9613b1949047569d,
    title = "Confidence Based Acoustic Event Detection",
    abstract = "Acoustic event detection, the determination of the acoustic event type and the localisation of the event, has been widely applied in many real-world applications. Many works adopt the multi-label classification technique to perform the polyphonic acoustic event detection with a global threshold to detect the active acoustic events. However, the manually labeled boundaries are error-prone and cannot always be accurate, especially when the frame length is too short to be accurately labeled by human annotators. To deal with this, a confidence is assigned to each frame and acoustic event detection is performed using a multi-variable regression approach in this paper. Experimental results on the latest TUT sound event 2017 database of polyphonic events demonstrate the superior performance of the proposed approach compared to the multi-label classification based AED method.",
    keywords = "Acoustic event detection, Confidence, Multi-label classification, Multi-variable regression",
    author = "Xianjun Xia and Roberto Togneri and Ferdous Sohel and David Huang",
    year = "2018",
    month = "9",
    day = "10",
    doi = "10.1109/ICASSP.2018.8461845",
    language = "English",
    isbn = "9781538646588",
    volume = "2018-April",
    pages = "306--310",
    booktitle = "2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    address = "United States",

    }

    Xia, X, Togneri, R, Sohel, F & Huang, D 2018, Confidence Based Acoustic Event Detection. in 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. vol. 2018-April, 8461845, IEEE, Institute of Electrical and Electronics Engineers, USA, pp. 306-310, 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018, Calgary, Canada, 15/04/18. https://doi.org/10.1109/ICASSP.2018.8461845

    Confidence Based Acoustic Event Detection. / Xia, Xianjun; Togneri, Roberto; Sohel, Ferdous; Huang, David.

    2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Vol. 2018-April USA : IEEE, Institute of Electrical and Electronics Engineers, 2018. p. 306-310 8461845.

    Research output: Chapter in Book/Conference paperConference paper

    TY - GEN

    T1 - Confidence Based Acoustic Event Detection

    AU - Xia, Xianjun

    AU - Togneri, Roberto

    AU - Sohel, Ferdous

    AU - Huang, David

    PY - 2018/9/10

    Y1 - 2018/9/10

    N2 - Acoustic event detection, the determination of the acoustic event type and the localisation of the event, has been widely applied in many real-world applications. Many works adopt the multi-label classification technique to perform the polyphonic acoustic event detection with a global threshold to detect the active acoustic events. However, the manually labeled boundaries are error-prone and cannot always be accurate, especially when the frame length is too short to be accurately labeled by human annotators. To deal with this, a confidence is assigned to each frame and acoustic event detection is performed using a multi-variable regression approach in this paper. Experimental results on the latest TUT sound event 2017 database of polyphonic events demonstrate the superior performance of the proposed approach compared to the multi-label classification based AED method.

    AB - Acoustic event detection, the determination of the acoustic event type and the localisation of the event, has been widely applied in many real-world applications. Many works adopt the multi-label classification technique to perform the polyphonic acoustic event detection with a global threshold to detect the active acoustic events. However, the manually labeled boundaries are error-prone and cannot always be accurate, especially when the frame length is too short to be accurately labeled by human annotators. To deal with this, a confidence is assigned to each frame and acoustic event detection is performed using a multi-variable regression approach in this paper. Experimental results on the latest TUT sound event 2017 database of polyphonic events demonstrate the superior performance of the proposed approach compared to the multi-label classification based AED method.

    KW - Acoustic event detection

    KW - Confidence

    KW - Multi-label classification

    KW - Multi-variable regression

    UR - http://www.scopus.com/inward/record.url?scp=85044675693&partnerID=8YFLogxK

    U2 - 10.1109/ICASSP.2018.8461845

    DO - 10.1109/ICASSP.2018.8461845

    M3 - Conference paper

    SN - 9781538646588

    VL - 2018-April

    SP - 306

    EP - 310

    BT - 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings

    PB - IEEE, Institute of Electrical and Electronics Engineers

    CY - USA

    ER -

    Xia X, Togneri R, Sohel F, Huang D. Confidence Based Acoustic Event Detection. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Vol. 2018-April. USA: IEEE, Institute of Electrical and Electronics Engineers. 2018. p. 306-310. 8461845 https://doi.org/10.1109/ICASSP.2018.8461845