Clever Toolbox - the Art of Automated Genre Classification

Peter Ahrendt (Author), Anders Meng (Author), Jan Larsen (Author), Sune Lehmann (Author)

    Research output: Non-textual formInteractive productionResearch

    Abstract

    Automatic musical genre classification can be defined as the science of finding computer algorithms that a digitized sound clip as input and yield a musical genre as output. The goal of automated genre classification is, of course, that the musical genre should agree with the human classificasion. This demo illustrates an approach to the problem that first extract frequency-based sound features followed by a "linear regression" classifier. The basic features are the so-called mel-frequency cepstral coefficients (MFCCs), which are extracted on a time-scale of 30 msec. From these MFCC features, auto-regressive coefficients (ARs) are extracted along with the mean and gain to get a single (30 dimensional) feature vector on the time-scale of 1 second. These features have been used because they have performed well in a previous study (Meng, Ahrendt, Larsen (2005)). Linear regression (or single-layer linear NN) is subsequently used for classification. This classifier is rather simple; current research investigates more advanced methods of classification.
    Original languageEnglish
    Publication date2005
    Place of PublicationKgs. Lyngby
    Publication statusPublished - 2005

    Cite this

    Ahrendt, P. (Author), Meng, A. (Author), Larsen, J. (Author), & Lehmann, S. (Author). (2005). Clever Toolbox - the Art of Automated Genre Classification. Interactive production, Kgs. Lyngby: Informatics and Mathematical Modelling, Technical University of Denmark.
    Ahrendt, Peter (Author) ; Meng, Anders (Author) ; Larsen, Jan (Author) ; Lehmann, Sune (Author). / Clever Toolbox - the Art of Automated Genre Classification. [Interactive production].
    @misc{d8910189edad431fa02dc065f3870938,
    title = "Clever Toolbox - the Art of Automated Genre Classification",
    abstract = "Automatic musical genre classification can be defined as the science of finding computer algorithms that a digitized sound clip as input and yield a musical genre as output. The goal of automated genre classification is, of course, that the musical genre should agree with the human classificasion. This demo illustrates an approach to the problem that first extract frequency-based sound features followed by a {"}linear regression{"} classifier. The basic features are the so-called mel-frequency cepstral coefficients (MFCCs), which are extracted on a time-scale of 30 msec. From these MFCC features, auto-regressive coefficients (ARs) are extracted along with the mean and gain to get a single (30 dimensional) feature vector on the time-scale of 1 second. These features have been used because they have performed well in a previous study (Meng, Ahrendt, Larsen (2005)). Linear regression (or single-layer linear NN) is subsequently used for classification. This classifier is rather simple; current research investigates more advanced methods of classification.",
    author = "Peter Ahrendt and Anders Meng and Jan Larsen and Sune Lehmann",
    year = "2005",
    language = "English",
    publisher = "Informatics and Mathematical Modelling, Technical University of Denmark",

    }

    Ahrendt, P, Meng, A, Larsen, J & Lehmann, S, Clever Toolbox - the Art of Automated Genre Classification, 2005, Interactive production, Informatics and Mathematical Modelling, Technical University of Denmark, Kgs. Lyngby.
    Clever Toolbox - the Art of Automated Genre Classification. Ahrendt, Peter (Author); Meng, Anders (Author); Larsen, Jan (Author); Lehmann, Sune (Author). 2005. Kgs. Lyngby : Informatics and Mathematical Modelling, Technical University of Denmark.

    Research output: Non-textual formInteractive productionResearch

    TY - ADVS

    T1 - Clever Toolbox - the Art of Automated Genre Classification

    A2 - Ahrendt, Peter

    A2 - Meng, Anders

    A2 - Larsen, Jan

    A2 - Lehmann, Sune

    PY - 2005

    Y1 - 2005

    N2 - Automatic musical genre classification can be defined as the science of finding computer algorithms that a digitized sound clip as input and yield a musical genre as output. The goal of automated genre classification is, of course, that the musical genre should agree with the human classificasion. This demo illustrates an approach to the problem that first extract frequency-based sound features followed by a "linear regression" classifier. The basic features are the so-called mel-frequency cepstral coefficients (MFCCs), which are extracted on a time-scale of 30 msec. From these MFCC features, auto-regressive coefficients (ARs) are extracted along with the mean and gain to get a single (30 dimensional) feature vector on the time-scale of 1 second. These features have been used because they have performed well in a previous study (Meng, Ahrendt, Larsen (2005)). Linear regression (or single-layer linear NN) is subsequently used for classification. This classifier is rather simple; current research investigates more advanced methods of classification.

    AB - Automatic musical genre classification can be defined as the science of finding computer algorithms that a digitized sound clip as input and yield a musical genre as output. The goal of automated genre classification is, of course, that the musical genre should agree with the human classificasion. This demo illustrates an approach to the problem that first extract frequency-based sound features followed by a "linear regression" classifier. The basic features are the so-called mel-frequency cepstral coefficients (MFCCs), which are extracted on a time-scale of 30 msec. From these MFCC features, auto-regressive coefficients (ARs) are extracted along with the mean and gain to get a single (30 dimensional) feature vector on the time-scale of 1 second. These features have been used because they have performed well in a previous study (Meng, Ahrendt, Larsen (2005)). Linear regression (or single-layer linear NN) is subsequently used for classification. This classifier is rather simple; current research investigates more advanced methods of classification.

    M3 - Interactive production

    PB - Informatics and Mathematical Modelling, Technical University of Denmark

    CY - Kgs. Lyngby

    ER -

    Ahrendt P (Author), Meng A (Author), Larsen J (Author), Lehmann S (Author). Clever Toolbox - the Art of Automated Genre Classification Kgs. Lyngby: Informatics and Mathematical Modelling, Technical University of Denmark. 2005.