Robust multipitch analyzer against initialization based on latent harmonic allocation using overtone corpus

Daichi Sakaue, Katsutoshi Itoyama, Tetsuya Ogata, Hiroshi G. Okuno

    Research output: Contribution to journalArticle

    Abstract

    We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed "overtone corpus." Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings.

    Original languageEnglish
    Pages (from-to)246-255
    Number of pages10
    JournalJournal of Information Processing
    Volume21
    Issue number2
    DOIs
    Publication statusPublished - 2013 Apr

    Fingerprint

    Musical instruments
    Frequency estimation

    Keywords

    • Harmonic clustering
    • Multipitch estimation
    • Musical instrument sounds
    • Overtone estimation

    ASJC Scopus subject areas

    • Computer Science(all)

    Cite this

    Robust multipitch analyzer against initialization based on latent harmonic allocation using overtone corpus. / Sakaue, Daichi; Itoyama, Katsutoshi; Ogata, Tetsuya; Okuno, Hiroshi G.

    In: Journal of Information Processing, Vol. 21, No. 2, 04.2013, p. 246-255.

    Research output: Contribution to journalArticle

    @article{d07c2fde04ad4d9bb522bda271c3c4ef,
    title = "Robust multipitch analyzer against initialization based on latent harmonic allocation using overtone corpus",
    abstract = "We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed {"}overtone corpus.{"} Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings.",
    keywords = "Harmonic clustering, Multipitch estimation, Musical instrument sounds, Overtone estimation",
    author = "Daichi Sakaue and Katsutoshi Itoyama and Tetsuya Ogata and Okuno, {Hiroshi G.}",
    year = "2013",
    month = "4",
    doi = "10.2197/ipsjjip.21.246",
    language = "English",
    volume = "21",
    pages = "246--255",
    journal = "Journal of Information Processing",
    issn = "0387-5806",
    publisher = "Information Processing Society of Japan",
    number = "2",

    }

    TY - JOUR

    T1 - Robust multipitch analyzer against initialization based on latent harmonic allocation using overtone corpus

    AU - Sakaue, Daichi

    AU - Itoyama, Katsutoshi

    AU - Ogata, Tetsuya

    AU - Okuno, Hiroshi G.

    PY - 2013/4

    Y1 - 2013/4

    N2 - We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed "overtone corpus." Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings.

    AB - We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed "overtone corpus." Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings.

    KW - Harmonic clustering

    KW - Multipitch estimation

    KW - Musical instrument sounds

    KW - Overtone estimation

    UR - http://www.scopus.com/inward/record.url?scp=84876222903&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84876222903&partnerID=8YFLogxK

    U2 - 10.2197/ipsjjip.21.246

    DO - 10.2197/ipsjjip.21.246

    M3 - Article

    AN - SCOPUS:84876222903

    VL - 21

    SP - 246

    EP - 255

    JO - Journal of Information Processing

    JF - Journal of Information Processing

    SN - 0387-5806

    IS - 2

    ER -