To main content

Synthetic Data For Dnn-Based Doa Estimation of Indoor Speech

Abstract

This paper investigates the use of different room impulse response (RIR) simulation methods for synthesizing training data for deep neural network-based direction of arrival (DOA) estimation of speech in reverberant rooms.

Different sets of synthetic RIRs are obtained using the image source method (ISM) and more advanced methods including diffuse reflections and/or source directivity. Multi-layer perceptron (MLP) deep neural network (DNN) models are trained on generalized cross correlation (GCC) features extracted for each set. Finally, models are tested on features obtained from measured RIRs.

This study shows the importance of training with RIRs from directive sources, as resultant DOA models achieved up to 51% error reduction compared to the steered response power with phase transform (SRP-PHAT) baseline (significant with p<<.01), while models trained with RIRs from omnidirectional sources did worse than the baseline. The performance difference was specifically present when estimating the azimuth of speakers not facing the array directly.

Category

Academic chapter/article/Conference paper

Client

  • Research Council of Norway (RCN) / 256753

Language

English

Author(s)

Affiliation

  • Norwegian University of Science and Technology
  • SINTEF Digital / Software and Service Innovation
  • SINTEF Digital / Smart Sensor Systems

Year

2021

Publisher

IEEE

Book

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Issue

2021

ISBN

978-1-7281-7606-2

Page(s)

4390 - 4394

View this publication at Cristin