Talker direction finder system implemented with cardioid beams and DSP algorithm

A prototype talker direction finder (TDF) algorithm has been developed that will rapidly locate a new talker (within 200–400 ms) or track a moving one. For most room configurations, the algorithm has an inaccuracy mean of less than 10° of azimuthal angle and a standard deviation of 7°. The TDF provi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of the Acoustical Society of America 1996-04, Vol.99 (4_Supplement), p.2504-2529
Hauptverfasser: Baumbauer, John C., Michel, Alan D., McAteer, Jeffrey P.
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A prototype talker direction finder (TDF) algorithm has been developed that will rapidly locate a new talker (within 200–400 ms) or track a moving one. For most room configurations, the algorithm has an inaccuracy mean of less than 10° of azimuthal angle and a standard deviation of 7°. The TDF provides full room, 360° coverage. A prototype where the TDF physically rotates a camera toward the talker has demonstrated the TDF’s robustness in difficult (i.e., realistic) noisy and reverberant environments. The physical inputs to the TDF are three small and inexpensive omnidirectional microphone elements that are all placed within a 1-in. circle, allowing inconspicuous design into a consumer product. Three μ-law codecs provide Inputs to the digital signal processor where the TDF’s new least-mean-squares-type adaptive algorithm resides. The overall system design is summarized, including preprocessing to reduce the effects of noise and reverberation. The adaptive algorithm, which employs redundant ‘‘pattern matching’’ to six cardioid acoustic pickup beams formed from the three microphone element inputs, is then presented. The TDF can provide enhanced solutions for teleconferencing and speakerphone acoustic systems. The algorithm could be used to provide directional information to steer an ‘‘adaptive beamformer’’ for advanced audio pickup applications.
ISSN:0001-4966
DOI:10.1121/1.415681