Classification of Acoustic Maps to Determine Speaker Position and Orientation from a Distributed Microphone Network

Abstract

Acoustic maps created on the basis of the signals acquired by distributed networks of microphones allow to identify position and orientation of an active talker in an enclosure. In adverse situations of high background noise, high reverberation or unavailability of direct paths to the microphones, localization may fail. This paper proposes a novel approach to talker localization and estimation of head orientation based on the classification of Global Coherence Field (GCF) or Oriented GCF maps. Preliminary experiments with data obtained by simulated propagation as well as with data acquired in a real room show that the match with precalculated map models provides a robust behavior in adverse conditions.

Automatic estimation of position and orientation of an acoustic source by a microphone array network

“…Six areas were randomly chosen inside the room, avoiding border areas. ͑Areas 1, 3,5,11,22,23,29,37,43,45,47, and 49 are border areas in Fig. 7.…”

“…Using the same standard metrics of the SRP-PHAT orientation method, i.e., the average error, the ability of the system to correctly classify the source orientation within eight classes separated by 45°, and assuming correct classification error of Ϯ1 adjacent class, 51°, 30%, and 68% were obtained, respectively. Brutti et al 19,23 extended the GCF position localization method to consider the source orientation. The new method was named the oriented global coherence field.…”

“…The source orientation [19][20][21][22][23] also plays an important role in acoustic localization because a directional source does not radiate uniformly in all directions, and the quality of signals recorded by distant microphones is affected not only by environmental noise and reverberation but also by the speaker's relative orientation. 21 Sachar et al 20 proposed the energy method, where differences in the source radiation pattern can be detected and used to predict the source orientation.…”

"Eye Array" Sound Source Localization

“…Some researchers have focused on using distributed arrays in SSL and defining some spatial related factors to the localization algorithm [29,30,31]. Distributed networks of microphones have also been used to create acoustic maps based on the classification of a global coherence field or oriented global coherence field to identify position and orientation of a speaker [32]. The near-and far-field arrays of such a distributed network is utilized in the European Commission integrated project CHIL, "Computers in the Human Interaction Loop" project, to solve the problems of speaker localization and tracking, speech activity detection and distant-talking automatic speech recognition [33].…”

Localization of multiple speakers based on a two step acoustic map analysis

“…Although more computationally expensive, it was shown that they provide reliable results. In particular, when even the maximization of the "global coherence" fails, a suitable analysis and classification of the spatial map yields useful information to localize a speaker and determine his/her head orientation [8].…”