Timezone: »
A typical audio signal processing pipeline includes multiple disjoint analysis stages, including calculation of a time-frequency representation followed by spectrogram-based feature analysis. We show how time-frequency analysis and nonnegative matrix factorisation can be jointly formulated as a spectral mixture Gaussian process model with nonstationary priors over the amplitude variance parameters. Further, we formulate this nonlinear model's state space representation, making it amenable to infinite-horizon Gaussian process regression with approximate inference via expectation propagation, which scales linearly in the number of time steps and quadratically in the state dimensionality. By doing so, we are able to process audio signals with hundreds of thousands of data points. We demonstrate, on various tasks with empirical data, how this inference scheme outperforms more standard techniques that rely on extended Kalman filtering.
Author Information
William Wilkinson (Queen Mary University of London)
Michael Andersen (Technical University of Denmark)
Joshua D. Reiss (Queen Mary University of London)
Dan Stowell (Queen Mary University of London)
Arno Solin (Aalto University)
Dr. Arno Solin is Assistant Professor in Machine Learning at the Department of Computer Science, Aalto University, Finland, and Adjunct Professor (Docent) at Tampere University, Finland. His research focuses on probabilistic models combining statistical machine learning and signal processing with applications in sensor fusion, robotics, computer vision, and online decision making. He has published around 50 peer-reviewed articles and one book. Previously, he has been a visiting researcher at Uppsala University (2019), University of Cambridge (2017-2018), and University of Sheffield (2014), and worked as a Team Lead in a tech startup. Prof. Solin is the winner of several prizes, hackathons, and modelling competitions, including the Schizophrenia Classification Challenge on Kaggle and the ISIF Jean-Pierre Le Cadre Best Paper Award. Homepage: http://arno.solin.fi
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Poster: End-to-End Probabilistic Inference for Nonstationary Audio Analysis »
Thu Jun 13th 01:30 -- 04:00 AM Room Pacific Ballroom
More from the Same Authors
-
2020 Poster: State Space Expectation Propagation: Efficient Inference Schemes for Temporal Gaussian Processes »
William Wilkinson · Paul Chang · Michael Andersen · Arno Solin -
2020 Poster: Scalable Exact Inference in Multi-Output Gaussian Processes »
Wessel Bruinsma · Eric Perim Martins · William Tebbutt · Scott Hosking · Arno Solin · Richard E Turner -
2020 Tutorial: Machine Learning with Signal Processing »
Arno Solin -
2019 Poster: Bayesian leave-one-out cross-validation for large data »
Måns Magnusson · Michael Andersen · Johan Jonasson · Aki Vehtari -
2019 Oral: Bayesian leave-one-out cross-validation for large data »
Måns Magnusson · Michael Andersen · Johan Jonasson · Aki Vehtari -
2018 Poster: State Space Gaussian Processes with Non-Gaussian Likelihood »
Hannes Nickisch · Arno Solin · Alexander Grigorevskiy -
2018 Oral: State Space Gaussian Processes with Non-Gaussian Likelihood »
Hannes Nickisch · Arno Solin · Alexander Grigorevskiy
