Audio signal separation through complex tensor factorization: Utilizing modulation frequency and phase information

Shogo Masaya

    Research output: Contribution to journalArticleScientificpeer-review

    4 Citations (Scopus)

    Abstract

    I propose a complex-valued tensor factorization algorithm for audio-source separation to exploit not only amplitude but phase information of audio signals in the modulation frequency (MF) domain. The proposed algorithm is extended from complex non-negative matrix factorization, which is capable of decomposing an arbitrary complex matrix such as the complex spectrum in the acoustic frequency domain. The proposed method enables us to factorize an arbitrary complex tensor of order 3. The detailed performance of the proposed algorithm for single-channel source separation is investigated through numerical experiments. I examine the quantitative contributions of the MF domain and phase information examined by additionally presenting three tensor factorization algorithms and using five objective indices for source separation.

    Original languageEnglish
    Pages (from-to)137-148
    JournalSignal Processing
    Volume142
    DOIs
    Publication statusPublished - 2018

    Keywords

    • Audio-source separation
    • Complex non-negative matrix factorization
    • Modulation frequency domain
    • Phase
    • Tensor factorization

    Fingerprint Dive into the research topics of 'Audio signal separation through complex tensor factorization: Utilizing modulation frequency and phase information'. Together they form a unique fingerprint.

    Cite this