Soft Computing

, Volume 16, Issue 12, pp 2027–2047 | Cite as

Multi-objective evolutionary feature selection for instrument recognition in polyphonic audio mixtures

  • Igor Vatolkin
  • Mike Preuß
  • Günter Rudolph
  • Markus Eichhoff
  • Claus Weihs
Original Paper

Abstract

Instrument recognition is one of the music information retrieval research topics. This task becomes very challenging if several instruments are played simultaneously because of their varying physical characteristics: inharmonic attack noise, energy development during attack–decay–sustain–release envelope or overtone distribution. In our framework, we treat instrument detection as a machine-learning task based on a large amount of preprocessed audio features with target to build classification models. Since classification algorithms are very sensitive to feature input and the optimal feature set differs from instrument to instrument, we propose to run a multi-objective feature selection procedure before building of classification models. Two objectives are considered for evaluation: classification mean-squared error and feature rate (smaller amount of features stands for reduced costs and decreased risk of overfitting). The analysis of the extensive experimental study confirms that application of an evolutionary multi-objective algorithm is a good choice to optimize feature selection for music instrument identification.

Keywords

Multi-objective feature selection Music classification Instrument recognition 

References

  1. Ahrendt P (2006) Music genre classification systems: Ph.D. thesis. Informatics and mathematical modelling, Technical University of DenmarkGoogle Scholar
  2. Beume N, Naujoks B, Emmerich M (2007) SMS-EMOA: multiobjective selection based on dominated hypervolume. Eur J Oper Res 181(3):1653–1669MATHCrossRefGoogle Scholar
  3. Bischl B, Mersmann O, Trautmann H, Weihs C (2012) Resampling methods for meta-model validation with recommendations for evolutionary computation. Evol Comput 20(2):249–275Google Scholar
  4. Bischl B, Vatolkin I, Preuß M (2010) Selecting small audio feature sets in music classification by means of asymmetric mutation. In: Proceedings of the 11th international conference on parallel problem solving from nature (PPSN). Springer, Berlin, pp 314–323Google Scholar
  5. Blume H, Bischl B, Botteck M, Igel C, Martin R, Rötter G, Rudolph G, Theimer W, Vatolkin I, Weihs C (2011) Huge music archives on mobile devices. IEEE Signal Process Mag 28(4):24–33CrossRefGoogle Scholar
  6. Brown JC, Houix O, McAdams S (2001) Feature dependence in the automatic identification of musical woodwind instruments. J Acoust Soc Am 109(3):1064–1072CrossRefGoogle Scholar
  7. Coello CAC, Van Veldhuizen DA, Lamont GB (eds) (2006) Evolutionary algorithms for solving multi-objective problems. Kluwer Academic Publishers, New YorkGoogle Scholar
  8. Deb K (2001) Multi-objective optimization using evolutionary algorithms: Wiley-Interscience Series in systems and optimization.. Wiley, ChichesterGoogle Scholar
  9. Eichhoff M, Weihs C (2012) Musical instrument recognition by high-level features. In: Gaul W, Geyer-Schulz A, Schmidt-Thieme L, Kunze J (eds) Challenges at the interface of data analysis, computer science, and optimization. Proceedings of the 34th annual conference of the Gesellschaft für Klassifikation e. V. Springer, Berlin, pp 373–381Google Scholar
  10. Emmanouilidis C, Hunter A, MacIntyre J (2000) A multiobjective evolutionary setting for feature selection and a commonality-based crossover operator. In: Proceedings of the 2000 Congress on Evolutionary Computation (CEC), vol 1. IEEE, pp 309–316Google Scholar
  11. Eronen A (2003) Musical instrument recognition using ica-based transform of features and discriminatively trained hmms. In: Proceedings of the 7th international symposium on signal processing and its applications (ISSPA), pp 133–136Google Scholar
  12. Eronen AJ (2009) Signal processing methods for audio classification and music content analysis: Ph.D. thesis. Department of Signal Processing, Tampere University of Technology, FinlandGoogle Scholar
  13. Essid S, Richard G, David B (2006) Instrument recognition in polyphonic music based on automatic taxonomies. IEEE Trans Audio Speech Lang Process 14(1):68–80CrossRefGoogle Scholar
  14. Fiebrink R, Fujinaga I (2006) Feature selection pitfalls and music classification. In: Proceedings of the 7th international conference on music information retrieval (ISMIR), pp 340–341Google Scholar
  15. Fu Z, Lu G, Ting K, Zhang D (2011) A survey of audio-based music classification and annotation. IEEE Trans Multimedia 13(2):303–319CrossRefGoogle Scholar
  16. Fujinaga I (1998) Machine recognition of timbre using steady-state tone of acoustic musical instruments. In: Proceedings of the international computer music conference (ICMC), ICMA, pp 207–210Google Scholar
  17. Goto M, Hashiguchi H, Nishimura T, Oka R (2003) Rwc music database: music genre database and musical instrument sound database. In: Proceedings of the 4th international conference on music information retrieval (ISMIR), pp 229–230Google Scholar
  18. Guyon I, Gunn S, Nikravesh M, Zadeh L (eds) (2006) Feature extraction, foundations and applications. Springer, BerlinGoogle Scholar
  19. Hall M (1999) Correlation-based feature selection for machine learning: Ph.D thesis. Department of Computer Science, Waikato University, New ZealandGoogle Scholar
  20. Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH (2009) The weka data mining software: an update. SIGKDD Explor 11:10–18CrossRefGoogle Scholar
  21. Heittola T, Klapuri A, Virtanen T (2009) Musical instrument recognition in polyphonic audio using source-filter model for sound separation. In: Proceedings of the 10th international society for music information retrieval conference (ISMIR), pp 327–332Google Scholar
  22. Kitahara T, Goto M, Komatani K, Ogata T, Okuno HG (2007) Instrument identification in polyphonic music: Feature weighting to minimize influence of sound overlaps. EURASIP J Adv Signal Process, vol 2007Google Scholar
  23. Kobayashi Y (2009) Automatic generation of musical instrument detector by using evolutionary learning method. In: Proceedings of the 10th international society for music information retrieval conference (ISMIR), pp 93–98Google Scholar
  24. Lartillot O, Toiviainen P (2007) Mir in Matlab (ii): a toolbox for musical feature extraction from audio. In: Proceedings of the 8th international conference on music information retrieval (ISMIR), pp 127–130Google Scholar
  25. Li T, Ogihara M, Tzanetakis G (2011) Music data mining. CRC Press, USAGoogle Scholar
  26. Liu J, Hu X (2010) User-centered music information retrieval evaluation. In: Proceedings of the joint conference on digital libraries (JCDL) workshop: music information retrieval for the massesGoogle Scholar
  27. Livshin A, Rodet X (2006) The significance of the non-harmonic “noise” versus the harmonic series for musical instrument recognition. In: Proceedings of the 7th international conference on music information retrieval (ISMIR), pp 95–100Google Scholar
  28. Loughrey J, Cunningham P (2004) Overfitting in wrapper-based feature subset selection: the harder you try the worse it gets. In: M. Bramer, F. Coenen, T. Allen (eds) Proceedings of the 24th SGAI international conference on innovative techniques and applications of artificial intelligence (AI-2004), pp. 33–43Google Scholar
  29. Lukashevich H (2008) Towards quantitative measures of evaluating song segmentation. In: Proceedings of the 9th international conference on music information retrival (ISMIR), pp 375–380Google Scholar
  30. McEnnis D, McKay C, Fujinaga I (2006) jAudio: additions and improvements. In: Proceedigs of the 7th international conference on music information retrieval (ISMIR), pp 385–386Google Scholar
  31. Mierswa I, Morik K (2005) Automatic feature extraction for classifying audio data. Mach Learn J 58(2–3):127–149MATHCrossRefGoogle Scholar
  32. Mierswa I, Wurst M, Klinkenberg R, Scholz M, Euler T (2006) Yale rapid prototyping for complex data mining tasks. In:Ungar L, Craven M, Gunopulos D, Eliassi-Rad T (eds) Proceedings of the 12th ACM SIGKDD international conference on knowledge discovery and data mining (KDD). ACM Press, New York, pp 935–940Google Scholar
  33. Müller M, Ewert S (2010) Towards timbre-invariant audio features for harmony-based music. IEEE Trans Audio Speech Lang Process 18(3):649–662CrossRefGoogle Scholar
  34. Müller M, Ewert S (2011) Chroma toolbox: Matlab implementations for extracting variants of chroma-based audio features. In: Proceedings of the 12th international conference on music information retrieval (ISMIR), pp 215–220Google Scholar
  35. Nagathil A, Göttel P, Martin R (2011) Hierarchical audio classification using cepstral modulation ratio regressions based on Legendre polynomials. In: Proceedings of the international conference on acoustics, speech and signal processing (ICASSP), Prague, Czech Republic, pp 2216–2219Google Scholar
  36. Park TH (2010) Introduction to digital signal processing: computer musically speaking. World Scientific, USAGoogle Scholar
  37. Quinlan JR (1993) C4.5: programs for machine learning. Morgan Kaufmann, Menlo ParkGoogle Scholar
  38. Ras ZW, Wieczorkowska A (2010) Advances in music information retrieval: studies in computational intelligence, vol 274. Springer, BerlinGoogle Scholar
  39. Reynolds A, Corne D, Chantler M (2010) Feature selection for multi-purpose predictive models: a many-objective task. In: Proceedings of the 11th international conference on parallel problem solving from nature (PPSN). Springer, Berlin, pp 384–393Google Scholar
  40. Rudolph G (2012) Evolutionary strategies. In: Rozenberg G, Bäck T, Kok J (eds) Handbook of natural computing. Springer, BerlinGoogle Scholar
  41. Snyman J (2005) Practical mathematical optimization: an introduction to basic optimization theory and classical and new gradient-based algorithms. Springer, BerlinGoogle Scholar
  42. Theimer W, Vatolkin I, Eronen A (2008) Definitions of audio features for music content description: technical report TR08-2-001. Chair of Algorithm Engineering, University of Dortmund, GermanyGoogle Scholar
  43. Vatolkin I (2012) Multi-objective evaluation of music classification. In: Gaul W, Geyer-Schulz A, Schmidt-Thieme L, Kunze L (eds) Challenges at the interface of data analysis, computer science, and optimization. Proceedings of the 34th annual conference of the Gesellschaft für Klassifikation e. V. Springer, Berlin, pp 401–410Google Scholar
  44. Vatolkin I, Preuß M, Rudolph G (2011) Multi-objective feature selection in music genre and style recognition tasks. In: Krasnogor N, Lanzi PN (eds) Proceedings of the 2011 genetic and evolutionary computation conference (GECCO). ACM Press, New York, pp 411–418Google Scholar
  45. Vatolkin I, Theimer W, Botteck M (2010) Amuse (advanced music explorer): a multitool framework for music data analysis. In: Downie JS, Veltkamp RC (eds) Proceedings of the 11th international society on music information retrieval conference (ISMIR), pp 33–38Google Scholar
  46. Vatolkin I, Theimer W, Rudolph G (2009) Design and comparison of different evolution strategies for feature selection and consolidation in music classification. In: Proceedings of the 2009 IEEE Congress on Evolutionary Computation (CEC). IEEE Press, pp 174–181Google Scholar
  47. Weihs C, Ligges U, Mörchen F, Müllensiefen D (2007) Classification in music research. Adv Data Anal Classif 1(3):255–291MathSciNetMATHCrossRefGoogle Scholar
  48. Zhu Z, Jia S, Ji Z (2010) Towards a memetic feature selection algorithm. IEEE Comput Intell Mag 5(2):41–53CrossRefGoogle Scholar
  49. Zitzler E, Brockhoff D, Thiele L (2007) The hypervolume indicator revisited: on the design of Pareto-compliant indicators via weighted integration. In: Proceedings of the conference on evolutionary multi-criterion optimization (EMO), vol 4403. Springer, Berlin, pp. 862–876Google Scholar

Copyright information

© Springer-Verlag 2012

Authors and Affiliations

  • Igor Vatolkin
    • 1
  • Mike Preuß
    • 1
  • Günter Rudolph
    • 1
  • Markus Eichhoff
    • 2
  • Claus Weihs
    • 2
  1. 1.Fakultät für Informatik Technische Universität DortmundDortmundGermany
  2. 2.Fakultät für Statistik Technische Universität DortmundDortmundGermany

Personalised recommendations