Joint speaker localization and array calibration using expectation-maximization

• Published in: EURASIP journal on audio, speech, and music processing Vol. 2020; no. 1; pp. 1 - 19
• Main Authors: Dorfan, Yuval, Schwartz, Ofer, Gannot, Sharon
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 09.06.2020; Springer Nature B.V; SpringerOpen
• Subjects: Acoustics; Ad hoc networks; Algorithms; Anchors; Arrays; Beamforming; Calibration; Computer simulation; Constellations; Electronic devices; Engineering; Engineering Acoustics; Expectation-maximization; Joint calibration and localization; Localization; Mathematics in Music; Maximization; Microphone array; Microphones; Optimization; Signal,Image and Speech Processing; Simultaneous speakers; W-disjoint; Wireless acoustic sensor network; Joint calibration and localization; Expectation-maximization; Microphone array; Wireless acoustic sensor network; W-disjoint; Simultaneous speakers
• ISSN: 1687-4722; 1687-4714; 1687-4722
• DOI: 10.1186/s13636-020-00177-1

Ad hoc acoustic networks comprising multiple nodes, each of which consists of several microphones, are addressed. From the ad hoc nature of the node constellation, microphone positions are unknown. Hence, typical tasks, such as localization, tracking, and beamforming, cannot be directly applied. To tackle this challenging joint multiple speaker localization and array calibration task, we propose a novel variant of the expectation-maximization (EM) algorithm. The coordinates of multiple arrays relative to an anchor array are blindly estimated using naturally uttered speech signals of multiple concurrent speakers. The speakers’ locations, relative to the anchor array, are also estimated. The inter-distances of the microphones in each array, as well their orientations, are assumed known, which is a reasonable assumption for many modern mobile devices (in outdoor and in a several indoor scenarios). The well-known initialization problem of the batch EM algorithm is circumvented by an incremental procedure, also derived here. The proposed algorithm is tested by an extensive simulation study.