公开(公告)号：US11373659B2

公开(公告)日：2022-06-28

申请号：US16713109

申请日：2019-12-13

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Kibeom Kim ,  Sunmin Kim ,  Hoon Heo ,  Inwoo Hwang ,  Kiwoong Kang

IPC: G10L19/00 ,  G06N3/04 ,  H04R3/00 ,  G09G3/22 ,  G06N3/08

Abstract: A display apparatus includes a display, an audio outputter, and a processor that divides, into data blocks, a data group having a predetermined number of frames with respect to data of an audio stream input frame by frame; sequentially performs operations using a filter previously learned according to the data blocks; identifies an audio feature according to the data group, based on the sequentially performed operations; and processes the data of the audio stream so that a sound output corresponding to the data group has sound effects corresponding to the identified audio feature.

公开(公告)号：US11462233B2

公开(公告)日：2022-10-04

申请号：US16685479

申请日：2019-11-15

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hoon Heo ,  Sunmin Kim ,  Kiwoong Kang ,  Kibeom Kim ,  Inwoo Hwang

IPC: G10L25/60 ,  G10L25/81 ,  G10L25/84 ,  G10L25/78 ,  H04R5/04 ,  G10L25/30

Abstract: An electronic device and method of recognizing an audio scene are provided. The method of recognizing an audio scene includes: separating, according to a predetermined criterion, an input audio signal into channels; recognizing, according to each of the separated channels, at least one audio scene from the input audio signal by using a plurality of neural networks trained to recognize an audio scene; and determining, based on a result of the recognizing of the at least one audio scene, at least one audio scene included in audio content by using a neural network trained to combine audio scene recognition results for respective channels, wherein the plurality of neural networks includes: a first neural network trained to recognize the audio scene based on a time-frequency shape of an audio signal, a second neural network trained to recognize the audio scene based on a shape of a spectral envelope of the audio signal, and a third neural network trained to recognize the audio scene based on a feature vector extracted from the audio signal.

公开(公告)号：US10991383B2

公开(公告)日：2021-04-27

申请号：US16713231

申请日：2019-12-13

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Inwoo Hwang ,  Sunmin Kim ,  Kiwoong Kang ,  Kibeom Kim ,  Hoon Heo

IPC: G10L25/51 ,  G06F16/22

Abstract: Provided are a method and a device for identifying a sound level of audio content, the method including receiving an audio signal and obtaining a first hash address based on a sound level of the audio signal, providing a hash address database in which the sound level of the audio signal is made to correspond to a hash address among a plurality of hash addresses, based on a first sound level corresponding to the first hash address and the sound level of the audio signal corresponding to the first sound level, cumulatively counting a first count value of the first hash address, and identifying the sound level of the audio content received during a first time period based on count information corresponding to a total amount of the first count value accumulated during the first time period.