Rhythm is it! - Discussion session at ISMIR next week

[Anja Volk]

Dear all,

thank you all for a very inspiring workshop last week! I want to transport some of the very interesting research questions that we have discussed to the computational guys at ISMIR - to spread the word about what can we do with computational methods that go beyond the typical MIR tasks of onset, beat, and tempo detection. Since ISMIR is taking place next week, I have taken the opportunity to initiate a discussion session there on the last day, and together with Andre and Ajay, who are also present at ISMIR, we hope to transport the sparks from Abu Dhabi to a discussion there (and anybody else from this list who will attend ISMIR, you are very welcome to join us!!).  

Our discussion session “Rhythm is it! MIR beyond onset, beat and tempo” is announced here at the ISMIR 2014 Unconference forum:

https://groups.google.com/a/ismir.net/forum/#!forum/ismir2014-unconference-forum

Please anyone who has ideas, contributions, comments - leave your posts there, it will enrich the preparation of the discussion session! And if you do not consider yourself a member of the MIR community (yet), see it as an opportunity to say "what I always dreamed MIR-researchers would explore about rhythm but never dared to ask” :-)

I attach the description of the discussion session as posted to the ISMIR Unconference forum at the end of this email.

Kind regards,

Anja

Discussion session, ISMIR Unconference 2014

Title: Rhythm is it! MIR beyond onset, beat and tempo.

Rhythm has been demonstrated to be a fundamental characteristic of music in many different contexts of listening and moving to, or creating of music. For instance, seven months old babies are able to distinguish different meters in music (Phillips-Silver & Trainor, 2005), melodies are best recalled when relying on rhythmic features (Sloboda & Parker, 1985; Schulkind et al., 2003) and differentiating between classical musical styles can be done by rhythm only (Dalla Bella & Peretz, 2005).

Yet, there are many open issues on rhythm as a fundamental feature of music and in different musical cultures that can benefit from computational data-rich approaches to music that MIR methods can provide. These go beyond the typical MIR tasks of beat tracking, onset detection, tempo estimation, or dance music classification (which are, without doubt, challenging tasks in themselves). For instance, in the recent “Cross-Disciplinary and Multi-Cultural Perspectives on Musical Rhythm and Improvisation” workshop at NYU Abu Dhabi (Oct 12-15, 2014) many open issues from rhythm research have been discussed, such as rhythmic patterns and underlying metric structures in Ottoman music, in north Indian rupak tal,  in polyrhythmic ensemble playing in Ghanaian dances, or challenges of rhythm in real-time composition systems and in evaluating rhythmic similarity measures. 

We want to use this discussion session as a platform for people working on rhythm in MIR to gather and discuss the following points: 

(1) What are currently typical approaches to rhythm in music in MIR? What research questions have been addressed in what ways, with what data, and with which methodology? What have we achieved?

(2) What research questions on rhythm in music have been overlooked so far in the MIR community that would profit from MIR methods?

(3) How do we envision to establish collaboration with people from musicology, music education, music cognition and other related fields to both feeding back into these areas as well as drawing inspiration from them for our research on rhythm?

References

Phillips-Silver & Trainor: Feeling the Beat: Movement Influences Infant Rhythm Perception, 2005

Sloboda & Parker:  Immediate recall of melodies, 1985; 

Schulkind, Posner, Rubin: Musical Features That Facilitate Melody Identification, 2003

Dalla Bella & Peretz: Differentiation of classical music requires little learning but rhythm, 2005

*********************************************************************************
Anja Volk, 
VIDI-laureate, Project leader MUSIVA 
Assistant Professor, MA, MSc, PhD

Department of Information and Computing Sciences, Utrecht University
PO Box 80.089
3508 TB Utrecht, the Netherlands
http://people.cs.uu.nl/volk/
Tel.:+31 (30) 253 5965
email: a.v...@uu.nl