We study the importance of a melodic audio (MA) feature set in music emotion recognition (MER) and compare its performance to an approach using only standard audio (SA) features. We also analyse the fusion of both types of features. Employing only SA features, the best attained performance was 46.3%, while using only MA features the best outcome was 59.1% (Fmeasure). A combination of SA and MA features improved results to 64%. These results might have an important impact to help break the so-called glass ceiling in MER, as most current approaches are based on SA features.
Subject
music emotion recognition
Related Project
MOODetector: A System for Mood-based Classification and Retrieval of Audio Music
Conference
6th International Workshop on Music and Machine Learning – MML’2013 – in conjunction with the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases – ECML/PKDD 2013, Prague, Czech Republic, September 2013
PDF File
Cited by
Year 2016 : 1 citations
Tomar D., Agarwal S. (2016) Multi-label Classifier for Emotion Recognition from Music. In: Nagar A., Mohapatra D., Chaki N. (eds) Proceedings of 3rd International Conference on Advanced Computing, Networking and Informatics. Smart Innovation, Systems and Technologies, vol 43. Springer, New Delhi
Year 2015 : 2 citations
Lee J, Kim D-W. Ranking Tag Pairs for Music Recommendation Using Acoustic Similarity. International Journal of Fuzzy Logic and Intelligent Systems. 2015;15(3):159-165. doi:10.5391/IJFIS.2015.15.3.159
Sundararajoo, Kohshelan (2015) "Improvement of audio feature extraction techniques in traditional Indian string musical instrument". Masters thesis, Universiti Tun Hussein Onn Malaysia.
