Editorial: Introducing the Transactions of the International Society for Music Information Retrieval

Abstract

Retrieval (TISMIR) publishes novel scientific research in the field of music information retrieval (MIR), an interdisciplinary research area concerned with processing, analysing, organising and accessing music information with computational methods. This definition suggests that the words information retrieval are not taken particularly literally in the ISMIR community, who have chosen to preserve the name for continuity. Meanwhile many ISMIR members think of MIR as music information research (Serra et al., 2013), or use an equivalent term such as music informatics. But academic pursuits do not confine themselves to disciplinary boundaries, and this is particularly true of the multidisciplinary and inter-disciplinary field of MIR. So TISMIR welcomes submissions from researchers working across a wide range of disciplines, including computer science, musicology, cognitive science, library & information science, acoustics and electrical engineering. Within this scope, topics of interest to TISMIR include:

The Anatomy of Consonance/Dissonance: Evaluating Acoustic and Cultural Predictors Across Multiple Datasets with Chords

“…For familiarity , the current model relies on corpus-based counts as an index of familiarity of the chords and intervals. This predictor (), implemented by Harrison and Pearce (2018), derives pitch-class frequencies from the Billboard corpus (Burgoyne, 2012) consisting of 739 pieces, which is probably currently the best source of information to represent common chords in Western popular music. This model is available in the library.…”

“…This predictor will be called as it is based on a corpus of popular music. We also carry out the same operation for the two other corpora available in (Harrison & Pearce, 2018), namely classical and jazz. The classical corpus ( n = 1,022) contains an assortment of Mozart, Chopin, Haydn, Bach, and Beethoven sonatas and string quartets ().…”

“…To be consistent with the other models of harmonicity, we invert the model output as it originally outputs high values for pitch combinations with high period length, which implies lower periodicity and consonance. Finally, Harrison and Pearce (2018) () have proposed a variant of Milne’s harmonicity model (Milne, 2013) where the template matching is not done with cosine distance but rather treating the profile as a probability distribution, which allows to measure the degree of the violation of the profile from the uniform distribution using Kullback-Leibler divergence.…”

“…This approach is also motivated by music theory, which suggests that inversions of a chord represent the same chord type as its root position (Rameau [1722], 1971). We re-encoded the Billboard corpus available in (Harrison & Pearce, 2018) as Forte classes using the conversion routine available in (Cuthbert & Ariza, 2010). Forte classes are defined by two numbers: the number of pitch classes, and the sequential number within that number of pitch-class sets.…”

Backbeat placement affects tempo judgments.

“…If a listener perceives the tempo to be halved or doubled, their preferred tactus would change to a metrical level once removed from the originally detected tactus. Octave errors have been revealed in numerous attempts at automatic tempo extraction in the music information retrieval community (Gkiokas et al, 2012; Hockman & Fujinaga, 2010; Hörschläger et al, 2015; Klapuri et al, 2006), but very little behavioral research has addressed this phenomenon directly.…”