Unifying Probabilistic Models for Time-Frequency Analysis

11/06/2018
by   William J. Wilkinson, et al.
0

In audio signal processing, probabilistic time-frequency models have many benefits over their non-probabilistic counterparts. They adapt to the incoming signal, quantify uncertainty, and measure correlation between the signal's amplitude and phase information, making time domain resynthesis straightforward. However, these models are still not widely used since they come at a high computational cost, and because they are formulated in such a way that it can be difficult to interpret all the modelling assumptions. By showing their equivalence to Spectral Mixture Gaussian processes, we illuminate the underlying model assumptions and provide a general framework for constructing more complex models that better approximate real-world signals. Our interpretation makes it intuitive to inspect, compare, and alter the models since all prior knowledge is encoded in the Gaussian process kernel functions. We utilise a state space representation to perform efficient inference via Kalman smoothing, and we demonstrate how our interpretation allows for efficient parameter learning in the frequency domain.

READ FULL TEXT

Authors

01/31/2019

End-to-End Probabilistic Inference for Nonstationary Audio Analysis

A typical audio signal processing pipeline includes multiple disjoint an...
02/02/2018

A Generative Model for Natural Sounds Based on Latent Force Modelling

Recent advances in analysis of subband amplitude envelopes of natural so...
11/17/2018

Designing nearly tight window for improving time-frequency masking

Many audio signal processing methods are formulated in the time-frequenc...
07/12/2020

State Space Expectation Propagation: Efficient Inference Schemes for Temporal Gaussian Processes

We formulate approximate Bayesian inference in non-conjugate temporal an...
11/09/2020

Bayesian Reconstruction of Fourier Pairs

In a number of data-driven applications such as detection of arrhythmia,...
07/27/2020

A Probabilistic Spectral Analysis of Multivariate Real-Valued Nonstationary Signals

A class of multivariate spectral representations for real-valued nonstat...
05/16/2022

Frequency selective extrapolation with residual filtering for image error concealment

The purpose of signal extrapolation is to estimate unknown signal parts ...

Code Repositories

unifying-prob-time-freq

Unifying probabilistic models for time-frequency analysis. Spectral Mixture Gaussian processes in state space form.


view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.