Sound emission and collection device and sound emission and collection method

摘要

Provided is a sound emission and collection device capable of estimating the azimuth of a sound source (such as a main utterer) precisely without any processing load. The sound emission and collection device (1) is connected with another sound emission and collection device via a network or the like. The sound emission and collection device (1) receives a sound signal from another sound emission and collection device, as a sound emission signal (FE), and emits the same from a speaker (SP). The sound emission and collection device (1) collects the sound at microphones (MIC1 to MIC3), and produces sound collection beam signals (NE1 to NE3) of different azimuths. The sound emission and collection device down-samples the individual sound collection beam signals (NE1 to NE3), and filters out the echoes of the down-sampled sound collection beam signals (DNE1 to DNE3). The sound emission and collection device selects the sound collection beam signal (DNE1′) of the highest signal level from the echo-filtered sound collection beam signals (DNE1′ to DNE3′). The sound emission and collection device filters out the echoes of a sound collection beam signal (NE1) from the sound collection azimuth (D1) of the sound collection beam signal (DNE1′), and transmits the same to another sound emission and collection device.

主权项

1. A sound emission and collection device comprising:
a sound collection section that collects sounds of a plurality of azimuths and produces sound collection signals for the azimuths respectively; a sound emission section that emits a sound; an azimuth estimation section that estimates an azimuth of a sound source based on signal levels of the sound collection signals for the azimuths respectively; a first echo cancel section that removes an echo of the sound collection signal from the azimuth of the sound source estimated by the azimuth estimation section; and a plurality of second echo cancel sections that remove echoes of the sound collection signals for the azimuths respectively, wherein the plurality of second echo cancel sections each process a smaller load than that of the first echo cancel section, and wherein the azimuth estimation section estimates the azimuth of the sound source based on signal levels of the sound collection signals having been subjected to an echo removal by the second echo cancel section.