This convention paper has been reproduced from the author’s advance manuscript, without editing, corrections, or consideration by the Review Board. The AES takes no responsibility for the contents. Additional papers

Abstract. Content-based music information retrieval uses features extracted from music to answer queries. For melodic queries, the two main features are the pitch and duration of notes. The note pitch feature has been well researched whereas duration has not been fully explored. In this paper, we discuss how the note duration feature can be used to alter music retrieval effectiveness. Notes are represented by strings called standardisations. A standardisation is designed for approximate string matching and may not capture melodic information precisely. To represent pitches, we use a string of pitch differences. Our duration standardisation uses a string of five symbols representing the relative durations of adjacent notes. For both features, the Smith-Waterman alignment is used for matching. We demonstrate combining the similarity in both features using a vector model. Results of our experiments in retrieval effectiveness show that note duration similarity by itself is not useful for effective music retrieval. Combining pitch and duration similarity using the vector model does not improve retrieval effectiveness over the use of pitch on its own. 1

In 1975 Parsons developed his dictionary of musical themes based on a simple contour representation. The motivation was that people with little training in music would be able to identify pieces of music. We decided to test whether people of various levels of musical skill could indeed make use of a text representation to describe a simple melody query. The results indicate that the task is beyond those who are unmusical, and that a scale numeric representation is easier than a contour one for those of moderate musical skill. Further, a common error when using the scale representation still yields a more accurate contour representation than if a user is asked to enter a contour query. We observed an average query length of about seven symbols for the retrieval task.

We have explored methods for music information retrieval for polyphonic music stored in the MIDI format. These methods use a query, expressed as a series of notes that are intended to represent a melody or theme, to identify similar pieces. Our work has shown that a three-phase architecture is appropriate for this task, in which the first phase is melody extraction, the second is standardisation, and the third is query-to-melody matching. We have investigated and systematically compared algorithms for each of these phases. To ensure that our results are robust, we have applied methodologies that are derived from text information retrieval: we developed test collections and compared different ways of acquiring test queries and relevance judgements. In this paper we review this program of work, compare to other approaches to music information retrieval, and identify outstanding issues.

Most research into music information retrieval thus far has only examined music from the western tradition. However, music of other origins often conforms to different tuning systems. Therefore there are problems both in representing this music as well as finding matches to queries from these diverse tuning systems. We discuss the issues associated with microtonal music retrieval and present some preliminary results from an experiment in applying scoring matrices to microtonal matching. 1.

In Query by Humming (QBH) systems the melody contour is often used as a symbolic description of music. The MelodyContour Description Scheme (DS) defined by MPEG-7 is a standardized representation of melody contours. For melody comparison in a QBH system a distance measure is required. This paper evaluates different distance measures for the MPEG7 MelodyContour DS. The usability of each measure is discussed. I.

Abstract: To date there has been very little research conducted on the behaviour of music information retrieval (MIR) users, in spite of the immense popularity of free music retrieval systems available on the Internet. In this study we examine the issue of music seeking behaviour through the examination of users life style effect of three different age groups using questionnaires. It was found that lifestyles had a significant impact on users need for music and hence their music seeking behaviour. The importance of social networks in music information seeking was reinforced in this study. An experiment was conducted with three different types of search on the Kazaa MIR system and the participants interviewed in order to collect data. Users found the Kazaa system intuitive and easy to use. Searchers used both song titles and lyrics for finding relevant music items. The insights provided by this study can be of assistance in the development of user focused Internet MIR systems.

The nearest neighbour problem is of practical significance in a number of fields. Often we are interested in finding an object near to a given query object. The problem is old, and a large number of solutions have been proposed for it in the literature. However, it remains the case that even the most popular of the techniques proposed for its solution have not been compared against each other. Also, many techniques, including the old and popular ones, can be implemented in a number of ways, and often the different implementations of a technique have not been thoroughly compared either. This research presents a detailed investigation of different implementations of two popular nearest neighbour search data structures, KDTrees and Metric Trees, and compares the different implementations of each of the two structures against each other. The best implementations of these structures are then compared against each other and against two other techniques, Annulus Method and Cover Trees. Annulus Method is an old technique that was rediscovered during the research for this thesis. Cover Trees are one of the most novel and promising data structures for nearest neighbour search that have been proposed in the literature. i Acknowledgments The continued support of Department of Computer Science’s Machine Learning group, and particularly my supervisor Dr. Eibe Frank, is greatly appreciated, without which this thesis would not have been possible.

Abstract Quite reasonable retrieval effectiveness is achieved for retrieving polyphonic (multiple notes at once) music that is symbolically encoded via melody queries, using relatively simple pattern matching techniques based on pitch sequences. Earlier work showed that adding duration information was not particularly helpful for improving retrieval effectiveness. In this paper we demonstrate that defining the duration information as the time interval between consecutive notes does lead to more effective retrieval when combined with pitch-based pattern matching in our collection of over 14 000 MIDI files.