Title:
METHOD AND DEVICE FOR PROCESSING AUDIO IN ORDER TO CLASSIFY SCENE
Document Type and Number:
WIPO Patent Application WO/2023/219292
Kind Code:
A1
Abstract:
A method for processing an audio according to an embodiment of the present disclosure may comprise the steps of: obtaining a first audio signal corresponding to a first frame; extracting a first feature vector by using a first neural network that uses the first audio signal as an input; obtaining a temporal correlation vector showing a similarity between the first feature vector and at least one second feature vector that has been extracted from at least one second audio signal corresponding to at least one second frame temporally preceding the first frame; and classifying a scene of the first audio signal by using a second neural network that uses the first feature vector, the at least one second feature vector, and the temporal correlation vector as an input.
Inventors:
KIM KYUNGRAE (KR)
NAM WOOHYUN (KR)
NAM WOOHYUN (KR)
Application Number:
PCT/KR2023/005182
Publication Date:
November 16, 2023
Filing Date:
April 17, 2023
Export Citation:
Assignee:
SAMSUNG ELECTRONICS CO LTD (KR)
International Classes:
G10L21/0272; G10L19/008; G10L21/038; G10L25/30
Foreign References:
| KR20190042730A | 2019-04-24 | |||
| KR20140017342A | 2014-02-11 | |||
| KR20200063290A | 2020-06-05 | |||
| KR20220005386A | 2022-01-13 |
Other References:
KO SANG-SUN, CHO HYE-SEUNG, KIM HYOUNG-GOOK: "Polyphonic sound event detection using multi-channel audio features and gated recurrent neural networks", THE JOURNAL OF THE ACOUSTICAL SOCIETY OF KOREA, THE ACOUSTICAL SOCIETY OF KOREA, SEOUL, vol. 36, no. 4, 31 December 2017 (2017-12-31), Seoul, pages 267 - 272, XP093006497, ISSN: 1225-4428, DOI: 10.7776/ASK.2017.36.4.267
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF: