2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

Technical Program

Paper Detail

Paper IDSAM-2.4
Paper Title SYNTHETIC DATA FOR DNN-BASED DOA ESTIMATION OF INDOOR SPEECH
Authors Femke B. Gelderblom, Norwegian University of Science and Technology & SINTEF, Norway; Yi Liu, Johannes Kvam, SINTEF, Norway; Tor Andre Myrvoll, Norwegian University of Science and Technology & SINTEF, Norway
SessionSAM-2: Direction of Arrival Estimation 2
LocationGather.Town
Session Time:Tuesday, 08 June, 16:30 - 17:15
Presentation Time:Tuesday, 08 June, 16:30 - 17:15
Presentation Poster
Topic Sensor Array and Multichannel Signal Processing: [SAM-MAPR] Microphone array processing
IEEE Xplore Open Preview  Click here to view in IEEE Xplore
Virtual Presentation  Click here to watch in the Virtual Conference
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.