Speaker localization and speech extraction with the ear sensor.

Julien Bonnal (1) , Sylvain Argentieri (1) , Patrick Danès (1) , Jérôme Manhes (2)

1 LAAS-RAP - Équipe Robotique, Action et Perception
2 LAAS-I2C - Service Instrumentation Conception Caractérisation

This paper presents the embedded audition for robotics (EAR) project internally developed at LAAS and its application to speaker localization and extraction. Hardware and software issues are first thoroughly depicted, concerning the development of an auditory sensor based on an array of microphones, a homemade dedicated acquisition chain and a FPGA based processing board. Then, the EAR sensor is assessed against various scenarios, in real noisy robotics environments. Localization results are presented when a speaker emits an utterance in the presence of a disturbing source. These validate the underlying theory and suggest further theoretical and experimental developments.

Domaines: Robotique [cs.RO]; Intelligence artificielle [cs.AI]; Traitement du signal et de l'image [eess.SP]

Julien Bonnal, Sylvain Argentieri, Patrick Danès, Jérôme Manhes. Speaker localization and speech extraction with the ear sensor.. IEEE/RSJ International Conference on Intelligent RObots and Systems (IROS), Oct 2009, St. Louis, MO, USA, United States. pp.670-675, ⟨10.1109/IROS.2009.5354401⟩. ⟨hal-03196430⟩