Advertisement

A Novel Singer Identification Method Using GMM-UBM

  • Xulong Zhang
  • Yiliang Jiang
  • Jin Deng
  • Juanjuan Li
  • Mi Tian
  • Wei LiEmail author
Conference paper
  • 211 Downloads
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 568)

Abstract

This paper presents a novel method for singer identification from polyphonic music audio signals. It is based on the universal background model (UBM), which is a singer-independent Gaussian mixture model (GMM) trained on many songs to model the singer characteristics. For our model, singing voice separation on a polyphonic signal is used to cope with the negative influences caused by background accompaniment. Then, we construct UBM for each singer trained with the Mel-frequency Cepstral Coefficients (MFCCs) feature, using the maximum a posterior (MAP) estimation. Singer identification is realized by matching test samples to the obtained UBMs for individual singers. Another major contribution of our work is to present two new large singer identification databases with over 100 singers. The proposed system is evaluated on two public datasets and two new ones. Results indicate that UBM can build more accurate statistical models of the singer’s voice than conventional methods. The evaluation carried out on the public dataset shows that our method achieves 16% improvement in accuracy compared with the state-of-the-art singer identification system.

Keywords

Singer identification (SI) Universal background model (UBM) Gaussian mixture model (GMM) 

References

  1. 1.
    Mesaros A (2013) Singing voice identification and lyrics transcription for music information re-trieval invited paper. In: Proceedings of IEEE 7th conference on speech technology and human-computer dialogue. Cluj-Napoca, Romania, pp 1–10Google Scholar
  2. 2.
    Hu Y, Liu G (2015) Separation of singing voice using nonnegative matrix partial co-factorization for singer identification. IEEE/ACM Trans Audio Speech Lang Process 23(4):643–653CrossRefGoogle Scholar
  3. 3.
    Yang YH (2013) Low-rank representation of both singing voice and music accompaniment via learned dictionaries. In: Proceedings of the international symposium on music information retrieval. Curitiba, Brazil, pp 427–432Google Scholar
  4. 4.
    Sha CY, Yang YH, Lin YC, Chen HH (2013) Singing voice timbre classification of Chinese popular music. In: Proceeding of IEEE international conference on acoustics, speech and signal processing. Vancouver, Canada, pp 734–7383)Google Scholar
  5. 5.
    Huang PS, Chen SD, Smaragdis P, Hasegawa-Johnson M (2012) Singing-voice separation from monaural recordings using robust principal component analysis. In: proceeding of IEEE international conference on acoustics, speech and signal processing. Kyoto, Japan, pp 57–60Google Scholar
  6. 6.
    Ezzaidi H, Bahoura M, Rouat J (2010) Singer and music discrimination based threshold in polyphonic music. In: Proceeding of IEEE international symposium on signal processing and information technology. Luxor, Egypt, pp 445–450Google Scholar
  7. 7.
    Kim YE, Whitman B (2002) Singer identification in popular music recordings using voice coding features. In: Proceedings of the 3rd international conference on music information retrieval. Baltimore, Maryland, pp 13–17Google Scholar
  8. 8.
    Logan B (2000) Mel frequency cepstral coefficients for music modeling. In: Proceedings of the international symposium on music information retrieval. Plymouth, Massachusetts, pp 1–11Google Scholar
  9. 9.
    Nakano T, Yoshii K, Goto M (2014) Vocal timbre analysis using latent Dirichlet allocation and cross-gender vocal timbre similarity. In: Proceeding of IEEE international conference on acoustics, speech and signal processing. Florence, Italy, pp 5202–5206Google Scholar
  10. 10.
    Zhang T (2003) Automatic singer identification. In: Proceedings of international conference on multimedia and expo. Baltimore, Maryland, pp 33–36Google Scholar
  11. 11.
    Fujihara H, Kitahara T, Goto M, Komatani K, Ogata T, Okuno HG (2005) Singer identification based on accompaniment sound reduction and reliable frame selection. In: Proceedings of the international symposium on music information retrieval. London, UK, pp 329–336Google Scholar
  12. 12.
    Eghbal-Zadeh H, Lehner B, Schedl M, Widmer G (2015) i-vectors for timbre-based music similarity and music artist classification. In: Proceedings of the international symposium on music information retrieval. Malaga, Spain, pp 554–560Google Scholar
  13. 13.
    Berenzweig A, Ellis DPW, Lawrence S (2002) Using voice segments to improve artist classification of music. In: Proceeding of 22nd international conference: virtual, synthetic, and entertainment audio. Espoo, Finland, pp 1–8Google Scholar
  14. 14.
    Hasan T, Hansen JH (2011) A study on universal background model training in speaker verification. IEEE Trans Audio Speech Lang Process 19(7):1890–1899CrossRefGoogle Scholar
  15. 15.
    Huang PS, Chen SD, Smaragdis P, Hasegawa-Johnson M (2012) Singing-voice separation from monaural recordings using robust principal component analysis. In: Proceeding of IEEE international conference on acoustics, speech and signal processing. Kyoto, Japan, pp 57–60Google Scholar
  16. 16.
    Charbuillet C, Tardieu D, Peeters G (2011) GMM-supervector for content-based music similarity. In: Proceeding of international conference on digital audio effects. Paris, France, pp 425–428Google Scholar
  17. 17.
    Hsu CL, Jang JSR (2010) On the improvement of singing voice separation for monaural recordings using the MIR-1 K dataset. IEEE Trans Audio Speech Lang Process 18(2):310–319CrossRefGoogle Scholar
  18. 18.
    Sadjadi SO, Slaney M, Heck L (2013) MSR identity toolbox-A MATLAB toolbox for speaker recognition research. Microsoft Research, Conversational Systems Research Center, pp 1–32Google Scholar
  19. 19.
    Ratanpara T, Patel N (2015) Singer identification using perceptual features and cepstral coefficients of an audio signal from Indian video songs. EURASIP J Audio Speech Music Process 2015(1):1–12CrossRefGoogle Scholar
  20. 20.
    Ellis DPW (2007) Classifying music audio with timbral and chroma features. In: Proceeding of the 8th international conference on music information retrieval. Vienna, Austria, pp 339–340Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Xulong Zhang
    • 1
  • Yiliang Jiang
    • 1
  • Jin Deng
    • 1
  • Juanjuan Li
    • 1
  • Mi Tian
    • 2
  • Wei Li
    • 1
    • 3
    Email author
  1. 1.School of Computer ScienceFudan UniversityShanghaiChina
  2. 2.Centre for Digital MusicQueen Mary University of LondonLondonUK
  3. 3.Shanghai Key Laboratory of Intelligent Information ProcessingFudan UniversityShanghaiChina

Personalised recommendations