ABSTRACT
In this study, we propose a method for multiphonic analysis using Non-Negative Matrix Factor 2-D Deconvolution (NMF2D) that has versatility and does not limit the number of instruments used in a music piece. This method solves the limitation of instrument by performing instrument estimation on the basis matrix decomposed by NMF2D. Experiments were conducted on a relatively simple piece of music with a short performance time. The instrumental estimation performance and the pitch estimation performance were not sufficient. Issues remain in the classification accuracy of the instrument estimation and the parameters of the Constant-Q transformation.
- J. C. Brown. 1990. Calculation of a constant Q spectral transform. Journal of the Acoustical Society of America. 89, 1 (Sept. 1990), 425–434. https://ci.nii.ac.jp/naid/20001708355/Google Scholar
- Kitamura Daichi, Saruwatari Hiroshi, Shikano Kiyohiro, Kondo Kazunobu, and T. Yu. 2013. Importance of Regularization in Superresolution-Based Multichannel Signal Separation with Nonnegative Matrix Factorization. Museon 2013, 99 (May 2013), 1–6.Google Scholar
- Hadrien Foroughmand and Geoffroy Peeters. 2018. Music retiler: Using NMF2D source separation for audio mosaicing. In Audio Mostly 2018 on Sound in Immersion and Emotion - AM’18. Association for Computing Machinery, Wrexham, United Kingdom, 1–7.Google Scholar
- Holger Kirchhoff, S. Dixon, and Anssi Klapuri. 2012. Multi-template shift-variant non-negative matrix deconvolution for semi-automatic music transcription. In Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012. IEEE, Kyoto, Japan, 415–420.Google Scholar
- Miron Kursa, Witold Rudnicki, Alicja Wieczorkowska, Elżbieta Kubera, and Agnieszka Kubik-Komar. 2009. Musical Instruments in Random Forest. In Foundations of Intelligent Systems. Springer Berlin Heidelberg, Prague, Czech Republic, 281–290.Google Scholar
- Seokjin Lee. 2020. Estimating the Rank of a Nonnegative Matrix Factorization Model for Automatic Music Transcription Based on Stein’s Unbiased Risk Estimator. Applied Sciences 10, 8 (April 2020), 1–19. https://doi.org/10.3390/app10082911Google Scholar
- Morten Mørup and Mikkel N. Schmidt. 2006. Sparse Non-negative Matrix Factor 2-D Deconvolution. Technical University of Denmark, Denmark.Google Scholar
- Hiroaki Nakajima, Daichi Kitamura, Norihiro Takamune, S. Koyama, H. Saruwatari, Nobutaka Ono, Y. Takahashi, and Kazunobu Kondo. 2016. Music signal separation using supervised NMF with all-pole-model-based discriminative basis deformation. 2016 24th European Signal Processing Conference (EUSIPCO) 24 (Aug. 2016), 1143–1147. https://doi.org/10.1109/EUSIPCO.2016.7760427Google ScholarCross Ref
- Aditya Nugraha, Antoine Liutkus, and Emmanuel Vincent. 2015. Multichannel Audio Source Separation With Deep Neural Networks. IEEE/ACM Transactions on Audio, Speech, and Language Processing 1 (June 2015), 1–13. https://doi.org/10.1109/TASLP.2016.2580946Google Scholar
- Bhathiya Rathnayake, K.M.K. Weerakoon, G.M.R.I. Godaliyadda, and M.P.B. Ekanayake. 2018. Toward Finding Optimal Source Dictionaries for Single Channel Music Source Separation Using Nonnegative Matrix Factorization. In 2018 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE, Bangalore, India, 1493–1500.Google Scholar
- Hiroshi Sawada, Nobutaka Ono, Hirokazu Kameoka, Daichi Kitamura, and Hiroshi Saruwatari. 2019. A review of blind source separation methods: Two converging routes to ILRMA originating from ICA and NMF. APSIPA Transactions on Signal and Information Processing 8 (Jan. 2019), 1–14. https://doi.org/10.1017/ATSIP.2019.5Google Scholar
- Mikkel N. Schmidt and Morten Mørup. 2006. Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation. In Independent Component Analysis and Blind Signal Separation(Lecture Notes in Computer Science, Vol. 3889), Justinian Rosca, Deniz Erdogmus, José C. Príncipe, and Simon Haykin (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 700–707. https://doi.org/10.1007/11679363_87Google ScholarDigital Library
- Paris Smaragdis and J. Brown. 2004. Non-negative Matrix Factor Deconvolution; Extraction of Multiple Sound Sources from Monophonic Inputs. In Independent Component Analysis and Blind Signal Separation. Springer Berlin Heidelberg, Berlin, Heidelberg, 494–499.Google Scholar
- P. Smaragdis and J. C. Brown. 2003. Non-negative matrix factorization for polyphonic music transcription. In 2003 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics. IEEE, New Paltz, NY, USA, 177–180.Google ScholarCross Ref
- Jordan B. L. Smith and M. Goto. 2018. Nonnegative Tensor Factorization for Source Separation of Loops in Audio. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 1 (April 2018), 171–175. https://doi.org/10.1109/ICASSP.2018.8461876Google ScholarDigital Library
- Fabian-Robert Stöter, Stefan Uhlich, Antoine Liutkus, and Yuki Mitsufuji. 2019. Open-Unmix - A Reference Implementation for Music Source Separation. Journal of Open Source Software 4 (Sept. 2019), 1667. https://doi.org/10.21105/joss.01667Google ScholarCross Ref
- Gino Angelo Velasco, Nicki Holighaus, Monika Doerfler, and Thomas Grill. 2011. Constructing an invertible constant-Q transform with nonstationary Gabor frames. In International Conference on Digital Audio Effects (DAFx 11). DAFx-11, Paris, France, DAFX1–DAFX7.Google Scholar
- Beiming Wang and Mark Plumbley. 2005. Musical audio stream separation by non-negative matrix factorization. In in Proc. UK Digital Music Research Network (DMRN) Summer Conf. Digital Music Research Network, Glasgow, Scotland, UK.Google Scholar
- F. Weninger, Jonathan Le Roux, John Hershey, and Shinji Watanabe. 2014. Discriminative NMF and its application to single-channel source separation. In Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, Singapore, Singapore, 865–869.Google Scholar
Index Terms
- Music Instrument Estimation and Multiple Sound Source Analysis from Monophonic inputs
Recommendations
Segregating Musical Chords for Automatic Music Transcription: A LSTM-RNN Approach
Pattern Recognition and Machine IntelligenceAbstractNotating or transcribing a music piece is very important for musicians. It not only helps them to communicate among each other but also helps in understanding a piece. This is very much essential for improvisations and performances. This makes ...
MOMOS-MT: mobile monophonic system for music transcription: sheet music generation on mobile devices
SAC '17: Proceedings of the Symposium on Applied ComputingMusic holds a significant cultural role in social identity and in the encouragement of socialization. Technology, by the destruction of physical and cultural distance, has lead to many changes in musical themes and the complete loss of forms. Yet, it ...
Monophonic constrained non-negative sparse coding using instrument models for audio separation and transcription of monophonic source-based polyphonic mixtures
In this paper we propose a monophonic constrained signal decomposition model applied to polyphonic signals composed of several monophonic sources from different musical instruments. The harmonic constraint is particularly effective for tonal instruments ...
Comments