On the Use of Beamforming Approaches for Binaural Speaker Localization

Konferenz: Speech Communication - 12. ITG-Fachtagung Sprachkommunikation
05.10.2016 - 07.10.2016 in Paderborn, Deutschland

Tagungsband: ITG-Fb. 267: Speech Communication

Seiten: 5Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Zohourian, Mehdi; Enzner, Gerald; Martin, Rainer (Institute of Communication Acoustics, Ruhr-Universität Bochum, Bochum, Germany)

This paper addresses the problem of binaural speaker localization while taking the rotational movement of the head into account. Some classical approaches in this context estimate the azimuth source position based on the heuristic combination of binaural cues. In localization with a microphone array in free field, however, optimizing a cost function is often used to estimate the direction-of-arrival (DOA). In this paper we present two novel binaural localization algorithms based on optimizing cost functions derived from beamforming techniques. Our methods are able to localize and track multiple source positions in the frontal horizontal plane without prior training of the interaction between source positions and binaural cues. The methods also enable the localization across a wide range of frequencies in the presence of reverberation and head movement. Results over various source/receiver positions demonstrate the efficiency of our algorithms.