A General Flexible Framework for the Handling of Prior Information in Audio Source Separation

By Alexey Ozerov, Emmanuel Vincent, Frédéric Bimbot. IEEE Transactions on Audio, Speech, and Language Processing, 2012

322 citations recorded by the source corpus.

Most audio source separation methods are developed for a particular scenario characterized by the number of sources and channels and the characteristics of the sources and the mixing process. In this paper, we introduce a general audio source separation framework based on a library of structured source models that enable the incorporation of prior knowledge about each source via user-specifiable constraints. While this framework generalizes several existing audio source separation methods, it also allows to imagine and implement new efficient methods that were not yet reported in the literature. We first introduce the framework by describing the model structure and constraints, explaining its generality, and summarizing its algorithmic implementation using a generalized expectation-maximization algorithm. Finally, we illustrate the above-mentioned capabilities of the framework by applying it in several new and existing configurations to different source separation problems. We have released a software tool named Flexible Audio Source Separation Toolbox (FASST) implementing a baseline version of the framework in Matlab.

Subfields: Signal Processing

View Paper: https://www.semanticscholar.org/paper/6d335f16b4d709306d5fe0361f9fedbf0ae50fc4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Signal Processing

Citations in the corpus

Nonnegative signal factorization with learnt instrument models for sound source separation in close-microphone recordings (2013, background): “3 Augmented NMF for parameter estimation The nonnegativity of the parameters is widely used in music transcription [12,16,17,22] and source separation [36,44].”
Uncertainty propagation through deep neural networks (2015, methodology): “First, the noisy signals have been enhanced using a multichannel NMF pre-processor [23, 24].”
Speech Enhancement with LSTM Recurrent Neural Networks and its Application to Noise-Robust ASR (2015, methodology): “It seems that 2ch-NMF (lower left corner) is an outlier, yet we believe this can be explained by the fact that it is not discriminatively trained (unlike the single-channel version used here).”
Automatic music transcription: challenges and future directions (2013, background): “separation, Ozerov et al. (2012) proposed a framework that enables the incorporation of prior knowledge about the number and types of sources, and the mixing model.”
Multichannel music separation with deep neural networks (2016, unknown): “This approach was successfully used up to some variants in [15]–[18], but never with DNN models for the sources.”
Towards Scaling Up Classification-Based Speech Separation (2013, background): “, [15], [36]), but statistical source models are usually required or need to be adapted.”
Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures (2014, methodology): “As potential further work in the speech enhancement domain, we could consider investigating the application of our framework to models with continuity constraints or factorial structure [19, 20, 21, 22].”
Open-Unmix - A Reference Implementation for Music Source Separation (2019, methodology): “In 2012, the Flexible Audio Source Separation Toolbox (FASST) was presented in (Ozerov, Vincent, & Bimbot, 2011; Salaün et al., 2014).”
A Consolidated Perspective on Multimicrophone Speech Enhancement and Source Separation (2017, unknown): “…parameters θ , but consists in computing directly the posterior distribution of the source STFT coefﬁcients while marginalizing over all possible model parameters: This leads to more accurate estimation, since the point estimate θ , as in the ML and MAP criteria (100), (104), is replaced by its…”