Random forest regression based acoustic event detection with bottleneck features

    Research output: Chapter in Book/Conference paperConference paper

    5 Citations (Scopus)

    Abstract

    This paper deals with random forest regression based acoustic event detection (AED) by combining acoustic features with bottleneck features (BN). The bottleneck features have a good reputation of being inherently discriminative in acoustic signal processing. To deal with the unstructured and complex real-world acoustic events, an acoustic event detection system is constructed using bottleneck features combined with acoustic features. Evaluations were carried out on the UPC-TALP and ITC-Irst databases which consist of highly variable acoustic events. Experimental results demonstrate the usefulness of the low-dimensional and discriminative bottleneck features with relative 5.33% and 5.51% decreases in error rates respectively.

    Original languageEnglish
    Title of host publication2017 IEEE International Conference on Multimedia and Expo, ICME 2017
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages157-162
    Number of pages6
    ISBN (Electronic)9781509060672
    DOIs
    Publication statusPublished - 28 Aug 2017
    Event2017 IEEE International Conference on Multimedia and Expo, ICME 2017 - Hong Kong, Hong Kong
    Duration: 10 Jul 201714 Jul 2017

    Conference

    Conference2017 IEEE International Conference on Multimedia and Expo, ICME 2017
    CountryHong Kong
    CityHong Kong
    Period10/07/1714/07/17

    Fingerprint

    Acoustics
    Acoustic signal processing

    Cite this

    Xia, X., Togneri, R., Sohel, F., & Huang, D. (2017). Random forest regression based acoustic event detection with bottleneck features. In 2017 IEEE International Conference on Multimedia and Expo, ICME 2017 (pp. 157-162). [8019418] IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICME.2017.8019418
    Xia, Xianjun ; Togneri, Roberto ; Sohel, Ferdous ; Huang, David. / Random forest regression based acoustic event detection with bottleneck features. 2017 IEEE International Conference on Multimedia and Expo, ICME 2017. IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 157-162
    @inproceedings{0fb2a963536e414bae84ff852d6a577f,
    title = "Random forest regression based acoustic event detection with bottleneck features",
    abstract = "This paper deals with random forest regression based acoustic event detection (AED) by combining acoustic features with bottleneck features (BN). The bottleneck features have a good reputation of being inherently discriminative in acoustic signal processing. To deal with the unstructured and complex real-world acoustic events, an acoustic event detection system is constructed using bottleneck features combined with acoustic features. Evaluations were carried out on the UPC-TALP and ITC-Irst databases which consist of highly variable acoustic events. Experimental results demonstrate the usefulness of the low-dimensional and discriminative bottleneck features with relative 5.33{\%} and 5.51{\%} decreases in error rates respectively.",
    keywords = "Acoustic event detection, Acoustic features, Bottleneck features, Random forest regression",
    author = "Xianjun Xia and Roberto Togneri and Ferdous Sohel and David Huang",
    year = "2017",
    month = "8",
    day = "28",
    doi = "10.1109/ICME.2017.8019418",
    language = "English",
    pages = "157--162",
    booktitle = "2017 IEEE International Conference on Multimedia and Expo, ICME 2017",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    address = "United States",

    }

    Xia, X, Togneri, R, Sohel, F & Huang, D 2017, Random forest regression based acoustic event detection with bottleneck features. in 2017 IEEE International Conference on Multimedia and Expo, ICME 2017., 8019418, IEEE, Institute of Electrical and Electronics Engineers, pp. 157-162, 2017 IEEE International Conference on Multimedia and Expo, ICME 2017, Hong Kong, Hong Kong, 10/07/17. https://doi.org/10.1109/ICME.2017.8019418

    Random forest regression based acoustic event detection with bottleneck features. / Xia, Xianjun; Togneri, Roberto; Sohel, Ferdous; Huang, David.

    2017 IEEE International Conference on Multimedia and Expo, ICME 2017. IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 157-162 8019418.

    Research output: Chapter in Book/Conference paperConference paper

    TY - GEN

    T1 - Random forest regression based acoustic event detection with bottleneck features

    AU - Xia, Xianjun

    AU - Togneri, Roberto

    AU - Sohel, Ferdous

    AU - Huang, David

    PY - 2017/8/28

    Y1 - 2017/8/28

    N2 - This paper deals with random forest regression based acoustic event detection (AED) by combining acoustic features with bottleneck features (BN). The bottleneck features have a good reputation of being inherently discriminative in acoustic signal processing. To deal with the unstructured and complex real-world acoustic events, an acoustic event detection system is constructed using bottleneck features combined with acoustic features. Evaluations were carried out on the UPC-TALP and ITC-Irst databases which consist of highly variable acoustic events. Experimental results demonstrate the usefulness of the low-dimensional and discriminative bottleneck features with relative 5.33% and 5.51% decreases in error rates respectively.

    AB - This paper deals with random forest regression based acoustic event detection (AED) by combining acoustic features with bottleneck features (BN). The bottleneck features have a good reputation of being inherently discriminative in acoustic signal processing. To deal with the unstructured and complex real-world acoustic events, an acoustic event detection system is constructed using bottleneck features combined with acoustic features. Evaluations were carried out on the UPC-TALP and ITC-Irst databases which consist of highly variable acoustic events. Experimental results demonstrate the usefulness of the low-dimensional and discriminative bottleneck features with relative 5.33% and 5.51% decreases in error rates respectively.

    KW - Acoustic event detection

    KW - Acoustic features

    KW - Bottleneck features

    KW - Random forest regression

    UR - http://www.scopus.com/inward/record.url?scp=85030261885&partnerID=8YFLogxK

    UR - http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=8014303

    U2 - 10.1109/ICME.2017.8019418

    DO - 10.1109/ICME.2017.8019418

    M3 - Conference paper

    SP - 157

    EP - 162

    BT - 2017 IEEE International Conference on Multimedia and Expo, ICME 2017

    PB - IEEE, Institute of Electrical and Electronics Engineers

    ER -

    Xia X, Togneri R, Sohel F, Huang D. Random forest regression based acoustic event detection with bottleneck features. In 2017 IEEE International Conference on Multimedia and Expo, ICME 2017. IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 157-162. 8019418 https://doi.org/10.1109/ICME.2017.8019418