Title:
電子機器および電子機器の制御方法
Document Type and Number:
Japanese Patent JP6661710
Kind Code:
B2
Abstract:
PROBLEM TO BE SOLVED: To provide electronic equipment that generates speaker identification information at each time on the basis of a time existence period of sound emitted by each sound source, and causes an object corresponding to an utterance time of each speaker as identified to be displayed on a screen.SOLUTION: Electronic equipment of an embodiment includes: frequency decomposing means for receiving digitized time-series amplitude data of two acoustic signals from a first microphone and a second microphone arranged at a predetermined distance, and generating a power value and a phase value at each frequency of the amplitude data in a time-series manner; section detection means for detecting a voice section on the basis of the result of the power value and the phase value; speech direction estimation means for detecting a speech direction of the voice section; speaker clustering means for generating speaker identification information at each time on the basis of a time existence period of sound emitted by each sound source; and user interface display processing means for causing an object corresponding to an utterance time of each speaker to be visually displayed on a display screen on the basis of the speaker identification information.SELECTED DRAWING: Figure 3
More Like This:
Inventors:
Fumitoshi Mizutani
Application Number:
JP2018146098A
Publication Date:
March 11, 2020
Filing Date:
August 02, 2018
Export Citation:
Assignee:
dynabook inc.
International Classes:
G10L17/00; G10L15/22; G10L15/28; G10L25/51; H04R1/40; H04R3/00
Domestic Patent References:
| JP2006254226A | ||||
| JP2010175614A | ||||
| JP2010054733A | ||||
| JP2013525848A |
Attorney, Agent or Firm:
Suzue International Patent Office