公开(公告)号：US20170365275A1

公开(公告)日：2017-12-21

申请号：US15545198

申请日：2015-09-23

Applicant: Samsung Electronics Co. Ltd. ,  Industry-University Cooperation Foundation Hanyang University

Inventor： Seung-yeol LEE ,  Joon-hyuk CHANG ,  Byeong-seob KO ,  Song-kyu PARK ,  Tae-jun PARK

IPC: G10L21/0364 ,  G10L21/02 ,  G10L15/06 ,  G10L15/30 ,  G10L25/84 ,  G10L21/0232

CPC classification number: G10L21/0364 ,  G10L15/063 ,  G10L15/30 ,  G10L21/0205 ,  G10L21/0208 ,  G10L21/0232 ,  G10L25/84

Abstract: Provided is a speech enhancement method and a system therefor. The speech enhancement method includes receiving at least one speech signal; generating a first speech signal by performing a primary speech enhancement on the at least one speech signal; selecting a noise removing gain corresponding to the first speech signal from pre-learned noise removing gain information; and generating a second speech signal by performing a secondary speech enhancement on the first speech signal based on the selected noise removing gain.

公开(公告)号：US20170287505A1

公开(公告)日：2017-10-05

申请号：US15507433

申请日：2015-09-03

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jae-hoon JEONG ,  Seung-yeol LEE ,  In-woo HWANG ,  Byeong-seob KO

IPC: G10L25/51 ,  G10L21/0232 ,  G06N99/00 ,  G10L15/10

CPC classification number: G10L25/51 ,  G06N20/00 ,  G10L15/10 ,  G10L21/0232

Abstract: Provided is a method for learning an audio signal. The method includes: acquiring at least one frequency-domain audio signal including frames; dividing the frequency-domain audio signal into at least one block by using a similarity between frames; acquiring a template vector corresponding to each block; acquiring a sequence of the acquired template vectors corresponding to at least one frame included in each block; and generating learning data including the acquired template vectors and the sequence of the template vectors.

公开(公告)号：US20200098385A1

公开(公告)日：2020-03-26

申请号：US16691813

申请日：2019-11-22

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jae-youn CHO ,  Weiwei CUI ,  Seung-yeol LEE

IPC: G10L25/78 ,  H04R1/32 ,  G10L21/0216 ,  G10L15/25 ,  G10L21/0208

Abstract: A speech enhancement method is provided. The speech enhancement method includes: estimating a direction of a speaker by using an input signal, generating direction information indicating the estimated direction, detecting speech of a speaker based on a result of the estimating the direction, and enhancing the speech of the speaker by using the direction information based on a result of the detecting the speech.

公开(公告)号：US20160150338A1

公开(公告)日：2016-05-26

申请号：US14896534

申请日：2014-06-05

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jin-sung KIM ,  Jae-hoon JEONG ,  Jung-eun PARK ,  Ki-wan EOM ,  Byeong-seob KO ,  Anant BAIJAL ,  In-woo HWANG ,  Sang-chul KO ,  Jeong-su KIM ,  Seung-yeol LEE ,  Jae-youn CHO

IPC: H04R29/00 ,  G10L25/72 ,  G10L25/18

CPC classification number: H04R29/008 ,  G08B1/08 ,  G08B13/1672 ,  G08B21/0208 ,  G10L25/18 ,  G10L25/72 ,  G10L25/78 ,  H04R29/004

Abstract: Provided is a method of operating an acoustic event detecting apparatus including: receiving an acoustic input; extracting frequency characteristics of the acoustic input; determining whether a first acoustic event has occurred by analyzing the extracted frequency characteristics; and in response to determining that the first acoustic event has occurred, acquiring data about at least one of a sound, an image, and a video from an environment outside of the acoustic event detecting apparatus and transmitting the data to a first device.

Abstract translation: 提供一种操作声学事件检测装置的方法，包括：接收声输入; 提取声输入的频率特性; 通过分析所提取的频率特性来确定是否发生了第一声学事件; 并且响应于确定已经发生第一声学事件，从声学事件检测装置外的环境获取关于声音，图像和视频中的至少一个的数据，并将数据发送到第一设备。

公开(公告)号：US20160111113A1

公开(公告)日：2016-04-21

申请号：US14895808

申请日：2014-05-30

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jae-youn CHO ,  Weiwei CUI ,  Seung-yeol LEE

IPC: G10L25/78 ,  G10L15/25 ,  G10L21/0216 ,  H04R1/32

CPC classification number: G10L25/78 ,  G10L15/25 ,  G10L21/0208 ,  G10L21/0216 ,  G10L2021/02166 ,  H04R1/326

Abstract: A speech enhancement method is provided. The speech enhancement method includes: estimating a direction of a speaker by using an input signal, generating direction information indicating the estimated direction, detecting speech of a speaker based on a result of the estimating the direction, and enhancing the speech of the speaker by using the direction information based on a result of the detecting the speech.

Abstract translation: 提供语音增强方法。 语音增强方法包括：通过使用输入信号来估计扬声器的方向，产生指示估计方向的方向信息，基于估计方向的结果来检测讲话者的语音，以及通过使用 基于检测语音的结果的方向信息。