Surrogate Source Model Learning for Determined Source Separation

by   Robin Scheibler, et al.

We propose to learn surrogate functions of universal speech priors for determined blind speech separation. Deep speech priors are highly desirable due to their high modelling power, but are not compatible with state-of-the-art independent vector analysis based on majorization-minimization (AuxIVA), since deriving the required surrogate function is not easy, nor always possible. Instead, we do away with exact majorization and directly approximate the surrogate. Taking advantage of iterative source steering (ISS) updates, we back propagate the permutation invariant separation loss through multiple iterations of AuxIVA. ISS lends itself well to this task due to its lower complexity and lack of matrix inversion. Experiments show large improvements in terms of scale invariant signal-to-distortion (SDR) ratio and word error rate compared to baseline methods. Training is done on two speakers mixtures and we experiment with two losses, SDR and coherence. We find that the learnt approximate surrogate generalizes well on mixtures of three and four speakers without any modification. We also demonstrate generalization to a different variation of the AuxIVA update equations. The SDR loss leads to fastest convergence in iterations, while coherence leads to the lowest word error rate (WER). We obtain as much as 36



There are no comments yet.


page 1

page 2

page 3

page 4

page 5


Convolutive Transfer Function Invariant SDR training criteria for Multi-Channel Reverberant Speech Separation

Time-domain training criteria have proven to be very effective for the s...

Monaural source separation: From anechoic to reverberant environments

Impressive progress in neural network-based single-channel speech source...

Single-Channel Speech Separation with Auxiliary Speaker Embeddings

We present a novel source separation model to decompose asingle-channel ...

Independent Vector Analysis via Log-Quadratically Penalized Quadratic Minimization

We propose a new algorithm for blind source separation of convolutive mi...

Independent Vector Analysis with Deep Neural Network Source Priors

This paper studies the density priors for independent vector analysis (I...

Heterogeneous Target Speech Separation

We introduce a new paradigm for single-channel target source separation ...
This week in AI

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