Tracking the odd: meter inference in musical audio using particle filters
In the field of MIR (Music Information Retrieval) the potential of Bayesian methods for signal analysis is widely acknowledged, but for many tasks in MIR Bayesian approaches are still the exception. In this talk a new state-of-the-art particle filter (PF) system is presented for the task of meter inference from a music audio signal. The system can be applied to determine the type of meter of a musical audio signal, and to track beats and downbeats in a unified framework. The new inference method is designed to overcome the problem of PFs in multi-modal state-spaces, which arise due to tempo and phase ambiguities in musical rhythm representations. Our recent work will be summarized in which we compare the new method with a hidden Markov model (HMM) system and several other PF schemes in terms of performance, speed and scalability on several audio datasets. We demonstrate that using the proposed system the computational complexity can be reduced drastically in comparison to the HMM while maintaining the same order of downbeat and beat tracking accuracy. Furthermore, our results on a culturally diverse collection will be illustrated, in which the proposed model is shown to infer the type of meter in a collection of music samples from India, Greece, and Turkey. Our results suggest that the proposed system is capable of meter inference in large culturally diverse music collections. We argue that the system can be easily adapted to musical styles, and therefore avoids inclusion of an ethnocentric bias into music recommendation and distribution software systems.
Andre Holzapfel is currently a post-doctoral researcher at Boğaziçi University, Istanbul, funded by a Marie-Curie IEF grant. In parallel he pursues his second doctoral degree in music at the Center for Advanced Music Studies (MIAM), Istanbul. Before his work in Istanbul he was working as a researcher for the CompMusic project (University Pompeu Fabra), for INESC TEC Porto, and for the Austrian Research Institute for Artificial Intelligence. He obtained his first Ph.D. in Computer Science at the University of Crete. His MIR related research focuses on models and inference schemes for the structure of music, with an emphasis on rhythm. His research in ethnomusicology focuses on music of Crete, and he investigates subjects of the interaction of music and technology. He is a regular performer of Greek Rembetiko music, with his main instruments the Turkish oud and the guitar. He currently directs a documentary movie on amateur Fado in the city of Porto. For further information, refer to www.rhythmos.org