Evaluating the Non-Intrusive Room Acoustics Algorithm with the ACE Challenge

TL;DR

This paper introduces a data-driven, single-channel method using neural networks to non-intrusively estimate room reverberation time and direct-to-reverberant ratio, evaluated through the ACE challenge.

Contribution

It proposes a novel neural network-based approach for non-intrusive acoustic parameter estimation, combining features and models to improve accuracy.

Findings

RMSD of 3.84 dB for DRR estimation

RMSD of 43.19% for T60 estimation

Effective combination of neural network estimates with SVM

Abstract

We present a single channel data driven method for non-intrusive estimation of full-band reverberation time and full-band direct-to-reverberant ratio. The method extracts a number of features from reverberant speech and builds a model using a recurrent neural network to estimate the reverberant acoustic parameters. We explore three configurations by including different data and also by combining the recurrent neural network estimates using a support vector machine. Our best method to estimate DRR provides a Root Mean Square Deviation (RMSD) of 3.84 dB and a RMSD of 43.19 % for T60 estimation.