DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks

Israel Mendoza Velazquez; Yi Ren; Yoichi Haneda; Hector Manuel Perez-Meana

doi:10.1109/GCCE53005.2021.9621359

DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks

Israel Mendoza Velazquez, Yi Ren, Yoichi Haneda, Hector Manuel Perez-Meana

Escuela Superior de Ingeniería Mecánica y Eléctrica (ESIME), Unidad Culhuacán

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

In this study, we explore a direction-of-arrival (DoA) estimation approach using the steered response power with phase transform (SRP-PHAT) and DeepSphere, a graph-based spherical convolutional neural network (CNN) suitable for spherical topology. The SRP-PHAT maps were adjusted to the Hierarchical Equal Area isoLatitude Pixelization (HEALPix) algorithm. We performed simulations for an Eigenmike spherical microphone array and different resolutions of the SRP-PHAT maps. Results show an improvement for the lower resolution maps, as the mean angular error for the Spherical CNN-derived maps was reduced by about half when compared with the original maps.

Original language	English
Title of host publication	2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	510-511
Number of pages	2
ISBN (Electronic)	9781665436762
DOIs	https://doi.org/10.1109/GCCE53005.2021.9621359
State	Published - 2021
Event	10th IEEE Global Conference on Consumer Electronics, GCCE 2021 - Kyoto, Japan Duration: 12 Oct 2021 → 15 Oct 2021

Publication series

Name	2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

Conference

Conference	10th IEEE Global Conference on Consumer Electronics, GCCE 2021
Country/Territory	Japan
City	Kyoto
Period	12/10/21 → 15/10/21

Keywords

DeepSphere
Direction-of-Arrival
Spherical Convolutional Neural Network
Spherical microphone array

Access to Document

10.1109/GCCE53005.2021.9621359

Cite this

Velazquez, I. M., Ren, Y., Haneda, Y., & Perez-Meana, H. M. (2021). DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks. In 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021 (pp. 510-511). (2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GCCE53005.2021.9621359

@inproceedings{30285e93584a4c3a85f79543ba7720db,

title = "DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks",

abstract = "In this study, we explore a direction-of-arrival (DoA) estimation approach using the steered response power with phase transform (SRP-PHAT) and DeepSphere, a graph-based spherical convolutional neural network (CNN) suitable for spherical topology. The SRP-PHAT maps were adjusted to the Hierarchical Equal Area isoLatitude Pixelization (HEALPix) algorithm. We performed simulations for an Eigenmike spherical microphone array and different resolutions of the SRP-PHAT maps. Results show an improvement for the lower resolution maps, as the mean angular error for the Spherical CNN-derived maps was reduced by about half when compared with the original maps.",

keywords = "DeepSphere, Direction-of-Arrival, Spherical Convolutional Neural Network, Spherical microphone array",

author = "Velazquez, {Israel Mendoza} and Yi Ren and Yoichi Haneda and Perez-Meana, {Hector Manuel}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 10th IEEE Global Conference on Consumer Electronics, GCCE 2021 ; Conference date: 12-10-2021 Through 15-10-2021",

year = "2021",

doi = "10.1109/GCCE53005.2021.9621359",

language = "Ingl{\'e}s",

series = "2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "510--511",

booktitle = "2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021",

address = "Estados Unidos",

}

Velazquez, IM, Ren, Y, Haneda, Y & Perez-Meana, HM 2021, DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks. in 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021. 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021, Institute of Electrical and Electronics Engineers Inc., pp. 510-511, 10th IEEE Global Conference on Consumer Electronics, GCCE 2021, Kyoto, Japan, 12/10/21. https://doi.org/10.1109/GCCE53005.2021.9621359

DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks. / Velazquez, Israel Mendoza; Ren, Yi; Haneda, Yoichi et al.
2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 510-511 (2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks

AU - Velazquez, Israel Mendoza

AU - Ren, Yi

AU - Haneda, Yoichi

AU - Perez-Meana, Hector Manuel

PY - 2021

Y1 - 2021

N2 - In this study, we explore a direction-of-arrival (DoA) estimation approach using the steered response power with phase transform (SRP-PHAT) and DeepSphere, a graph-based spherical convolutional neural network (CNN) suitable for spherical topology. The SRP-PHAT maps were adjusted to the Hierarchical Equal Area isoLatitude Pixelization (HEALPix) algorithm. We performed simulations for an Eigenmike spherical microphone array and different resolutions of the SRP-PHAT maps. Results show an improvement for the lower resolution maps, as the mean angular error for the Spherical CNN-derived maps was reduced by about half when compared with the original maps.

AB - In this study, we explore a direction-of-arrival (DoA) estimation approach using the steered response power with phase transform (SRP-PHAT) and DeepSphere, a graph-based spherical convolutional neural network (CNN) suitable for spherical topology. The SRP-PHAT maps were adjusted to the Hierarchical Equal Area isoLatitude Pixelization (HEALPix) algorithm. We performed simulations for an Eigenmike spherical microphone array and different resolutions of the SRP-PHAT maps. Results show an improvement for the lower resolution maps, as the mean angular error for the Spherical CNN-derived maps was reduced by about half when compared with the original maps.

KW - DeepSphere

KW - Direction-of-Arrival

KW - Spherical Convolutional Neural Network

KW - Spherical microphone array

UR - http://www.scopus.com/inward/record.url?scp=85123481777&partnerID=8YFLogxK

U2 - 10.1109/GCCE53005.2021.9621359

DO - 10.1109/GCCE53005.2021.9621359

M3 - Contribución a la conferencia

AN - SCOPUS:85123481777

T3 - 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

SP - 510

EP - 511

BT - 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th IEEE Global Conference on Consumer Electronics, GCCE 2021

Y2 - 12 October 2021 through 15 October 2021

ER -

Velazquez IM, Ren Y, Haneda Y, Perez-Meana HM. DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks. In 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 510-511. (2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021). doi: 10.1109/GCCE53005.2021.9621359

DOA Estimation for Spherical Microphone Array using Spherical Convolutional Neural Networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this