公开(公告)号：US20190253823A1

公开(公告)日：2019-08-15

申请号：US15984490

申请日：2018-05-21

Applicant: Electronics and Telecommunications Research Institute

Inventor： Dae Young JANG ,  Yong Ju LEE ,  Jae-hyoun YOO ,  Sang Won SUH

IPC: H04S7/00

CPC classification number: H04S7/30 ,  H04S2400/01

Abstract: Provided is a stereophonic sound reproducing apparatus that applies a multi-rendering scheme to a channel sound signal and an object sound signal to enhance a stereophonic effect. A stereophonic sound reproducing method performed by the stereophonic sound reproducing apparatus may include receiving a channel sound signal based on a channel, an object sound signal based on an object, and metadata, and reproducing the channel sound signal based on a preset rendering scheme and reproducing, based on the metadata including a rendering scheme determined using the object sound signal, each object sound signal using the determined rendering scheme.

公开(公告)号：US20170163978A1

公开(公告)日：2017-06-08

申请号：US15228333

申请日：2016-08-04

Applicant: Electronics and Telecommunications Research Institute

Inventor： Mi Suk LEE ,  Kyeong Ok KANG ,  Tae Jin PARK ,  Seung Kwon BEACK ,  Sang Won SUH ,  Jong Mo SUNG ,  Tae Jin LEE

IPC: H04N17/00 ,  H04N21/43 ,  H04N21/8547 ,  H04N19/467 ,  H04N19/577 ,  H04N19/44 ,  H04N21/8358 ,  H04N19/593 ,  H04N19/52 ,  H04N21/242 ,  H04N19/85

CPC classification number: H04N17/004 ,  H04N19/44 ,  H04N19/467 ,  H04N19/52 ,  H04N19/577 ,  H04N19/593 ,  H04N19/85 ,  H04N21/242 ,  H04N21/4305 ,  H04N21/4307 ,  H04N21/4341 ,  H04N21/4348 ,  H04N21/6336 ,  H04N21/8358 ,  H04N21/8547

Abstract: A system and method for synchronizing an audio signal and a video signal are provided. A decoding method in the system may include decoding an audio signal and a video signal received from an encoding apparatus, extracting first unique information of the audio signal from the decoded video signal, generating second unique information of the audio signal based on the decoded audio signal, determining a delay between the audio signal and the video signal by comparing the first unique information to the second unique information, and synchronizing the audio signal and the video signal based on the delay. The first unique information may be generated based on an audio signal that is not encoded by the encoding apparatus, and may be inserted into the video signal.

公开(公告)号：US20200302949A1

公开(公告)日：2020-09-24

申请号：US16562110

申请日：2019-09-05

Applicant: Electronics and Telecommunications Research Institute

Inventor： Young Ho JEONG ,  Sang Won SUH ,  Tae Jin LEE ,  Woo-taek LIM ,  Hui Yong KIM

IPC: G10L25/30 ,  G10L15/14 ,  G10L17/04 ,  G10L25/84 ,  G06N3/04

Abstract: Provided is a sound event recognition method that may improve a sound event recognition performance using a correlation between difference sound signal feature parameters based on a neural network, in detail, that may extract a sound signal feature parameter from a sound signal including a sound event, and recognize the sound event included in the sound signal by applying a convolutional neural network (CNN) trained using the sound signal feature parameter.

公开(公告)号：US20200302917A1

公开(公告)日：2020-09-24

申请号：US16657605

申请日：2019-10-18

Applicant: Electronics and Telecommunications Research Institute ,  GANGNEUNG-WONJU NATIONAL UNIVERSITY INDUSTRY ACADEMY COOPERATION GROUP

Inventor： Young Ho JEONG ,  Sang Won SUH ,  Woo-taek LIM ,  SUNG WOOK PARK ,  HYEON GI MOON ,  YOUNG CHEOL PARK ,  SHIN HYUK JEON

IPC: G10L15/16 ,  G10L25/03 ,  G06N3/08 ,  G06N20/00 ,  G06F17/16 ,  G06F17/27

Abstract: A data augmentation method includes extracting one or more basis vectors and coefficient vectors corresponding to sound source data classified in advance into a target class by applying non-negative matrix factorization (NMF) to the sound source data, generating a new basis vector using the extracted basis vectors, and generating new sound source data using the generated new basis vector and the extracted coefficient vectors.

公开(公告)号：US20210174252A1

公开(公告)日：2021-06-10

申请号：US16927691

申请日：2020-07-13

Applicant: Electronics and Telecommunications Research Institute ,  Kyungpook National University Industry-Academic Cooperation Foundation

Inventor： Young Ho JEONG ,  Soo Young PARK ,  Sang Won SUH ,  Woo-taek LIM ,  Minhan KIM ,  Seokjin LEE

IPC: G06N20/00 ,  G06F17/18

Abstract: Disclosed is an apparatus and method for augmenting training data using a notch filter. The method may include obtaining original data, and obtaining training data having a modified frequency component from the original data by filtering the original data using a filter configured to remove a component of a predetermined frequency band.

公开(公告)号：US20200312350A1

公开(公告)日：2020-10-01

申请号：US16561824

申请日：2019-09-05

Applicant: Electronics and Telecommunications Research Institute

Inventor： Woo-taek LIM ,  Sang Won SUH ,  Young Ho JEONG

IPC: G10L25/30 ,  G10L19/26 ,  G10L15/32 ,  G10L15/10 ,  G10L15/02

Abstract: A sound event detection method includes receiving a sound signal and determining and outputting whether a sound event is present in the sound signal by applying a trained neural network to the received sound signal, and performing post-processing of the output to reduce an error in the determination, wherein the neural network is trained to early stop at an optimal epoch based on a different threshold for each of at least one sound event present in a pre-processed sound signal. That is, the sound event detection method may detect an optimal epoch to stop training by applying different characteristics for respective sound events and improve the sound event detection performance based on the optimal epoch.