Computational auditory induction as a missing-data model-fitting problem with Bregman divergence

The human auditory system has the ability, known as auditory induction, to estimate the missing parts of a continuous auditory stream briefly covered by noise and perceptually resynthesize them. In this article, we formulate this ability as a model-based spectrogram analysis and clustering problem with missing data, show how to solve it using an auxiliary function method, and explain how this method is generally related to the expectation–maximization (EM) algorithm for a certain type of divergence measures called Bregman divergences, thus enabling the use of prior distributions on the parameters. We illustrate how our method can be used to simultaneously analyze a scene and estimate missing information with two algorithms: the first, based on non-negative matrix factorization (NMF), performs analysis of polyphonic multi-instrumental musical pieces. Our method allows this algorithm to cope with gaps within the audio data, estimating the timbre of the instruments and their pitch, and reconstructing the missing parts. The second, based on a recently introduced technique for the analysis of complex acoustical scenes called harmonic-temporal clustering (HTC), enables us to perform robust fundamental frequency estimation from incomplete speech data.

The human auditory system has the ability, known as auditory induction, to estimate the missing parts of a continuous auditory stream briefly covered by noise and perceptually resynthesize them. In this article, we formulate this ability as a model-based spectrogram analysis and clustering problem with missing data, show how to solve it using an auxiliary function method, and explain how this method can be generally related to the EM algorithm for a certain type of divergence measures called Bregman divergences, enabling the use of prior distributions on the parameters. We illustrate how our method can be used to simultaneously analyze a scene and estimate missing information into it through the development of two algorithms: the first, based on non-negative matrix factorization (NMF), enables us to analyze a polyphonic multi-instrumental music piece in spite of the presence of gaps, estimating the timber of the instruments, their activations’ time and pitch, and reconstructing the missing parts; the second, based on a recently introduced technique for the analysis of complex acoustical scenes called Harmonic-Temporal Clustering (HTC), enables us to perform robust fundamental frequency estimation of speech on incomplete data.Figure optionsDownload as PowerPoint slideResearch highlights
► Auditory induction as a model-based spectrogram analysis problem with missing data.
► Simultaneously analyze acoustic scenes and estimate missing information.
► Link between missing-data model fitting and the EM algorithm for Bregman divergences.
► Robust F0 estimation of incomplete speech data using Harmonic-Temporal Clustering.