Benjamin Martin
ABSTRACT
With the growing presence of large collections of musical content, methods for facilitating efficient browsing and fast comparisons of audio pieces become more and more useful. Notably, methods that isolate relevant parts in audio pieces give an insight of the musical content and can be used to improve similarity evaluation systems. In this context, we propose an indexing method that allows retrieving in audio signals particular parts, namely a major repetition. We use harmonic representations together with string matching techniques to strictly define and isolate such segments. Experiments on state-of-the-art structural datasets show a strong correlation between the retrieved parts and the perceived structure of pieces.
- M. A. Bartsch and G. H. Wakefield. Audio thumbnailing of popular music using chroma-based representations. IEEE Transactions on Multimedia, 7(1):96--104, 2005. Google Scholar
Digital Library
- W. Chai. Automated analysis of musical structure. PhD thesis, Massachusetts Institute of Technology, Cambridge, MA, USA, pp 39--62, 2005. Google ScholarDigital Library
- M. Cooper and J. Foote. Summarizing popular music via structural similarity analysis. In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics., pp 127--130, 2004.Google Scholar
- J. S. Downie. Music information retrieval. Annual Review of Information Science and Technology, 37:295--340, 2003.Google ScholarCross Ref
- M. Goto. A chorus-section detecting method for musical audio signals. In Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), volume 5, pp 437--440, 2003.Google Scholar
- S. Kannan and E. W. Myers. An algorithm for locating non-overlapping regions of maximum alignment score. In CPM '93: Proceedings of the 4th Annual Symposium on Combinatorial Pattern Matching, pp 74--86, London, UK, 1993. Springer-Verlag. Google ScholarDigital Library
- M. Levy and M. Sandler. Structural segmentation of musical audio by constrained clustering. IEEE Transactions on Audio, Speech and Language Processing, 16(2):318--326, 2008. Google ScholarDigital Library
- H. M. Lukashevich. Towards quantitative measures of evaluating song segmentation. In Proc. of the 9th International Conference on Music Information Retrieval (ISMIR), pp 375--380, 2008.Google Scholar
- M. Mauch et al. Omras2 metadata project 2009. In 10th International Conference on Music Information Retrieval (ISMIR), Late-Breaking Session, Kobe, Japan, 2009.Google Scholar
- R. Middleton. "Form", Key Terms in Popular Music and Culture. Wiley-Blackwell, 1999.Google Scholar
- W. Miller. An algorithm for locating a repeated region, unpublished manuscript, 1992.Google Scholar
- M. Müller. Information retrieval for music and motion. Springer-Verlag New York Inc, 2007. Google ScholarDigital Library
- S. B. Needleman and C. D. Wunsch. A general method applicable to the search for similarities in the amino acid sequence of two proteins. Journal of Molecular Biology, 48(3):443--453, 1970.Google ScholarCross Ref
- J. Paulus, M. Müller, and A. Klapuri. Audio-based music structure analysis. In Proc. of the 11th International Conference on Music Information Retrieval (ISMIR), pp 625--636, Utrecht, Netherlands, 2010.Google Scholar
- J. Serrà et al. Chroma binary similarity and local alignment applied to cover song identification. IEEE Transactions on Audio, Speech and Language Processing, 16:1138--1151, 2008. Google ScholarDigital Library
- T. F. Smith and M. S. Waterman. Identification of common molecular subsequences. Journal of molecular biology, 147(1):195--197, March 1981.Google Scholar
Index Terms
- Indexing musical pieces using their major repetition
Recommendations
Precise pitch profile feature extraction from musical audio for key detection
The majority of pieces of music, including classical and popular music,are composed using music scales, such as keys. The key or the scale information of a piece provides important clues on its high level musical content, like harmonic and melodic ...
Cloud-smart Musical Instrument Interactions: Querying a Large Music Collection with a Smart Guitar
Large online music databases under Creative Commons licenses are rarely recorded by well-known artists, therefore conventional metadata-based search is insufficient in their adaptation to instrument players’ needs. The emerging class of smart musical ...
Analysis of Recognition of a Musical Instrument in Sound Mixes Using Support Vector Machines
Experiments with recognition of the dominating musical instrument in sound mixes are interesting from the point of view of music information retrieval, but this task can be very difficult if the mixed sounds are of the same pitch. In this paper, we ...
Reviews
Soubhik Chakraborty
Music information retrieval (MIR) techniques that can detect crucial parts in a musical piece help us understand the musical content of the piece, and improve the similarity evaluation systems. In this paper, the authors have developed a novel automatic indexing mechanism of musical sequences that retrieves the major repetition therein. The strategy is to extract this main repetition from the sequences in such a way that it "corresponds to two (non-overlapping) parts of the piece that share the highest degree of similarity."? In other words, this indexing technique effectively picks up the longest repetitive portion in the musical piece. Basically, their algorithm is just another string-matching technique. They report some success with their technique in the identification of a cover song. However, this assertion by the authors took me by surprise: "Music[al] pieces feature a strong redundancy. ... However, few works focus on the redundant aspect of music for musical indexing purpose[s]."? I think that it is well known that musical pieces do contain repetition. Even in Indian classical music, which claims to be extempore, there are typical note combinations that characterize a raga, which are repeated in a performance. Moreover, the significance of a melody, which we may define as a sequence of notes that is complete in some musical sense (and hence, can be taken as a single entity), can be calculated simply by multiplying the length of the melody (which is the number of notes in it) by the number of times the melody occurs in such a monophonic (single melody line) musical piece; it is evident that both the melody length and the frequency of occurrence are important in assessing its significance. A more complicated formula exists in polyphonic music (multiple melody lines), but, even there, melody lengths as well as repetitions are taken into account [1]. The paper is not very technical; apart from researchers, undergraduates and post-graduates who are interested in music information retrieval should be able to understand the content. Online Computing Reviews Service
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments