Title:
VOICE SIGNAL PROCESSING DEVICE, VOICE SIGNAL PROCESSING METHOD, AND VOICE SIGNAL PROCESSING PROGRAM
Document Type and Number:
WIPO Patent Application WO/2023/149254
Kind Code:
A1
Abstract:
A voice signal processing device (1) comprises: a speaker detection unit (102) that detects a near-end speaker included in an image captured by a camera (2); a speaker direction specification unit (103) that specifies a direction in which the near-end speaker is located; a voice arrival direction detection unit (105) that detects an arrival direction of a voice on the basis of a voice signal collected by a microphone array (3); an utterance inference unit (106) that infers whether or not the near-end speaker is uttering on the basis of the direction in which the near-end speaker is located and the arrival direction of the voice; an echo suppressor (107) that sets an amplification factor of the voice signal to a value of 1 or more when the near-end speaker is inferred to be uttering, that sets an amplification factor of the voice signal to a value of less than 1 when the near-end speaker is inferred not to be uttering, and that adjusts the level of the voice signal on the basis of the set amplification factor; and a voice output unit (109) that outputs the adjusted voice signal as a transmission signal to be transmitted to a far-end speaker.
More Like This:
Inventors:
HIROSE YOSHIFUMI (JP)
Application Number:
PCT/JP2023/001863
Publication Date:
August 10, 2023
Filing Date:
January 23, 2023
Export Citation:
Assignee:
PANASONIC IP CORP AMERICA (US)
International Classes:
H04R3/00; G10L21/0208; G10L21/034; G10L25/51; G10L25/84; H04R1/40; H04R3/02
Domestic Patent References:
| WO2019239667A1 | 2019-12-19 |
Foreign References:
| JP2012216998A | 2012-11-08 | |||
| JP2019075748A | 2019-05-16 |
Attorney, Agent or Firm:
KOTANI, Masataka et al. (JP)
Download PDF: