Speech enhancement with mixture-of-deep-experts with clean clustering pre-training

TL;DR

This paper introduces a mixture-of-deep-experts neural network architecture for single microphone speech enhancement, utilizing specialized experts and a gating mechanism to improve robustness and reduce complexity.

Contribution

The novel MoDE architecture employs multiple expert DNNs with a gating network for improved speech enhancement and noise robustness, with clean clustering pre-training enhancing performance.

Findings

Enhanced speech quality with MoDE architecture

Improved robustness to unfamiliar noise types

Reduced computational complexity during testing

Abstract

In this study we present a mixture of deep experts (MoDE) neural-network architecture for single microphone speech enhancement. Our architecture comprises a set of deep neural networks (DNNs), each of which is an 'expert' in a different speech spectral pattern such as phoneme. A gating DNN is responsible for the latent variables which are the weights assigned to each expert's output given a speech segment. The experts estimate a mask from the noisy input and the final mask is then obtained as a weighted average of the experts' estimates, with the weights determined by the gating DNN. A soft spectral attenuation, based on the estimated mask, is then applied to enhance the noisy speech signal. As a byproduct, we gain reduction at the complexity in test time. We show that the experts specialization allows better robustness to unfamiliar noise types.