专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明公开

KR1020010096678A 멀티채널 스피커에서 실시간으로 크로스톡을 제거하여 입체음향을 재생하는 시스템 有权
标题翻译：通过实时移除多通道扬声器中的十字路口再现3D声音的方法和系统
公开(公告)号：KR1020010096678A
公开(公告)日：2001-11-08
申请号：KR1020010041323
申请日：2001-07-10
申请人： 주식회사 이머시스
发明人： 김풍민 , 김현석 , 금종룡 , 배희연 , 이균 , 이동우 , 이진성 , 최용석
IPC分类号： H04S3/00
摘要： PURPOSE: A method and system for reproducing a 3D sound by removing cross-talk in a multichannel speaker in real time is provided to improve a sound position effect, enhance an environmental effect, realize an optimized 3D sound and drastically improve product quality and reliability. CONSTITUTION: A head related transfer function database(13) gathers a head related transfer function in a database when a speaker(1-7) and a listener(30) are located within a standard radius by channel. A location information unit(21) detects a location information with an angle, a distance between the listener(30) and the multichannel speaker(23) when a sound source(11) is inputted. A spatial information unit(12) removes a cross-talk phenomenon of the speakers(23) positioned within a preset area by channel by using the head related transfer function database(13) and the stored spatial information. A signal distributer(22) distributes signals to the speakers after the location information unit(21) obtains the speaker information when the sound source(11) is transferred to the signal distributer(22) through a cross-talk remover(14).
摘要翻译：目的：提供一种通过实时消除多声道扬声器中的串扰来再现3D声音的方法和系统，以改善声音位置效果，增强环境效果，实现优化的3D声音，并显着提高产品质量和可靠性。构成：当扬声器（1-7）和收听者（30）位于标准半径范围内时，头部相关传输函数数据库（13）在数据库中收集头部相关传输函数。当输入声源（11）时，位置信息单元（21）检测具有角度，收听者（30）和多声道扬声器（23）之间的距离的位置信息。空间信息单元（12）通过使用头部相关传输函数数据库（13）和存储的空间信息来消除位于预设区域内的扬声器（23）的串扰现象。当通过串扰去除器（14）将声源（11）传送到信号分配器（22）时，信号分配器（22）在位置信息单元（21）获得扬声器信息之后，向扬声器分配信号。

2. 发明授权

KR100849027B1 음성 신호에 대한 립싱크 동기화 방법 및 장치 失效
标题翻译：嘴唇同步到语音信号的同步方法和装置
公开(公告)号：KR100849027B1
公开(公告)日：2008-07-29
申请号：KR1020080001103
申请日：2008-01-04
申请人： 주식회사 이머시스
发明人： 김풍민 , 김현석 , 신동선 , 금종룡 , 김종우
IPC分类号： G10L15/28 , G10L15/22 , G10L15/24 , G06T17/00
摘要： 본 발명은 음성신호를 아바타의 립싱크에 동기화하기 위한 데이터 처리 방법 및 장치에 관한 것으로서, 입력되는 음성 신호를 처리하여 보다 사실적인 립싱크가 이루어지게 한다.
본 발명은, 입력되는 음성신호를 처리하여 립싱크 동기화를 하기 위한 데이터 처리 방법에 있어서, 입력 장치를 통해 입력된 입력 신호에 대해 주변 환경으로부터 입력되는 배경 노이즈를 저감시켜서 음성신호를 찾는 노이즈저감 및 음성인식 단계; 상기 음성신호에서 모음 부분을 추출하는 모음정보 추출 단계; 상기 모음 정보를 이용하여 상기 모음의 발성시의 입술의 움직임에 대한 정보를 추출하는 입술모션 추출단계; 상기 입술의 움직임 정보를 립싱크용 데이터로 변환하는 입술 모션정보 인코딩 단계를 포함하는 것을 특징으로 한다.
본 발명을 이용하면 음성과 관련된 별도의 텍스트 데이터를 입력받지 않고 실시간으로 입력된 음성 신호만을 이용하여 아바타에 좀 더 사실감 있는 립싱크를 제공하는 것이 가능하여 영상통화시 사용하는 대체화면의 립싱크에 적용할 경우 음성과 입모양이 어색하게 되는 것을 방지할 수 있어 통화 시 만족도를 높일 수 있다.
음성, 립싱크, 노이즈 저감, 영상통화, 아바타

3. 发明授权

KR100443405B1 멀티채널 스피커용 오디오 신호를 멀티 채널 헤드폰용 오디오 신호로 변환하여 재분배 하는 장치 有权
标题翻译： 멀티채널스피커용오디오신호를멀티채널헤드폰용오디오신호로변환하여재분배하는장치
公开(公告)号：KR100443405B1
公开(公告)日：2004-08-09
申请号：KR1020010040109
申请日：2001-07-05
申请人： 주식회사 이머시스
发明人： 김풍민 , 김현석 , 금종룡 , 배희연 , 이균 , 이동우 , 이진성 , 최용석
IPC分类号： H04S5/02
摘要： PURPOSE: A method and apparatus for converting an audio signal for a multichannel speaker into an audio signal for a multichannel headphone and redistributing the same is provided to effectively listen to a 3D sound from a headphone at lower costs, drastically enlarge applications, and improve product quality and reliability. CONSTITUTION: A plurality of speakers divide sounds felt in both ears of a listener into signals reaching to the both ears of the listener by right-and-left signal dividing filters. Signal redistributing filters convert the right and left signals divided from the speakers into audio signals for a plurality of headphones. The signals of the speakers are redistributed into the plurality of signals for headphones by using the signal redistributing systems composed of a hardware and a software. A 3D sound which is felt only in existing multichannel speaker systems can be effectively heard through headphones in a simple manner. Better 3D sound effect can be achieved at lower costs. Thus, application fields can be drastically enlarged. By virtue of the construction stated above, product quality and reliability are considerably improved.
摘要翻译：目的：提供一种将多声道扬声器的音频信号转换为多声道耳机的音频信号并重新分配的方法和装置，以便以更低的成本有效地从头戴式耳机收听3D声音，大幅扩大应用范围并改进产品质量和可靠性。构成：多个扬声器通过左右信号分配滤波器将收听者两耳中感觉到的声音分成到达收听者双耳的信号。信号重新分配滤波器将从扬声器分出的右和左信号转换成多个耳机的音频信号。通过使用由硬件和软件组成的信号重新分配系统，将扬声器的信号重新分配到耳机的多个信号中。只有在现有的多声道扬声器系统中才能感受到的3D声音可以通过耳机以简单的方式有效地听到。更低的成本可以实现更好的3D音效。因此，应用领域可以大大扩大。凭借上述结构，产品质量和可靠性得到显着改善。

4. 发明公开

KR1020010096676A 멀티채널 스피커용 오디오 신호를 멀티 채널 헤드폰용 오디오 신호로 변환하여 재분배 하는 장치 有权
标题翻译：用于将多声道扬声器的音频信号转换成用于多声道耳机的音频信号并重新分配它们的方法和装置
公开(公告)号：KR1020010096676A
公开(公告)日：2001-11-08
申请号：KR1020010040109
申请日：2001-07-05
申请人： 주식회사 이머시스
发明人： 김풍민 , 김현석 , 금종룡 , 배희연 , 이균 , 이동우 , 이진성 , 최용석
IPC分类号： H04S5/02
摘要： PURPOSE: A method and apparatus for converting an audio signal for a multichannel speaker into an audio signal for a multichannel headphone and redistributing the same is provided to effectively listen to a 3D sound from a headphone at lower costs, drastically enlarge applications, and improve product quality and reliability. CONSTITUTION: A plurality of speakers divide sounds felt in both ears of a listener into signals reaching to the both ears of the listener by right-and-left signal dividing filters. Signal redistributing filters convert the right and left signals divided from the speakers into audio signals for a plurality of headphones. The signals of the speakers are redistributed into the plurality of signals for headphones by using the signal redistributing systems composed of a hardware and a software. A 3D sound which is felt only in existing multichannel speaker systems can be effectively heard through headphones in a simple manner. Better 3D sound effect can be achieved at lower costs. Thus, application fields can be drastically enlarged. By virtue of the construction stated above, product quality and reliability are considerably improved.
摘要翻译：目的：提供一种用于将多声道扬声器的音频信号转换为多声道耳机的音频信号并重新分配其的方法和装置，以便以更低的成本有效地收听来自耳机的3D声音，大大地增加了应用程序，并且改进了产品质量和可靠性。构成：多个扬声器将收听者的两只耳朵中感觉到的声音分成通过左右信号分频滤波器到达听众双耳的信号。信号再分配滤波器将从扬声器分离的左右信号转换成多个耳机的音频信号。通过使用由硬件和软件组成的信号重新分配系统，将扬声器的信号重新分配到用于耳机的多个信号中。只能在现有的多声道扬声器系统中感觉到的3D声音可以通过耳机以简单的方式被有效地听到。可以以更低的成本实现更好的3D音效。因此，应用领域可以大幅扩大。凭借上述结构，产品质量和可靠性大大提高。

5. 发明公开

KR1020080008432A 음성 신호에 대한 립싱크 동기화 방법 및 장치 失效
标题翻译： LIP同步到语音信号的同步方法和装置
公开(公告)号：KR1020080008432A
公开(公告)日：2008-01-23
申请号：KR1020080001103
申请日：2008-01-04
申请人： 주식회사 이머시스
发明人： 김풍민 , 김현석 , 신동선 , 금종룡 , 김종우
IPC分类号： G10L15/28 , G10L15/22 , G10L15/24 , G06T17/00
摘要： A method and an apparatus for lip-sync synchronization for a voice signal are provided to offer realistic lip-sync to an avatar by using only an inputted voice signal in real-time without receiving separate text data about the voice signal. A data processing method for lip-sync synchronization comprises the following steps of: detecting a voice signal by reducing background noise inputted from a surrounding environment for an input signal through an input device(1); extracting a vowel part of the voice signal(2); extracting information about the motion of lips in speaking the vowel part by using the vowel information(3); and converting the motion information about lips into lip-sync data(4).
摘要翻译：提供了一种用于语音信号的唇同步同步的方法和装置，以通过仅使用输入的语音信号实时地向化身提供现实的唇同步，而不接收关于语音信号的单独的文本数据。用于唇同步同步的数据处理方法包括以下步骤：通过减少通过输入装置（1）输入信号的从周围环境输入的背景噪声来检测语音信号; 提取语音信号（2）的元音部分; 通过使用元音信息（3）提取关于嘴唇部分的元音动词的信息; 并将关于嘴唇的运动信息转换成唇同步数据（4）。

6. 发明授权

KR100443404B1 멀티채널 스피커에서 실시간으로 크로스톡을 제거하여 입체음향을 재생하는 시스템 有权
标题翻译：在本实用新型的实施方式中，本发明的实施方式旨在提供一种方法，
公开(公告)号：KR100443404B1
公开(公告)日：2004-08-09
申请号：KR1020010041323
申请日：2001-07-10
申请人： 주식회사 이머시스
发明人： 김풍민 , 김현석 , 금종룡 , 배희연 , 이균 , 이동우 , 이진성 , 최용석
IPC分类号： H04S3/00
摘要： PURPOSE: A method and system for reproducing a 3D sound by removing cross-talk in a multichannel speaker in real time is provided to improve a sound position effect, enhance an environmental effect, realize an optimized 3D sound and drastically improve product quality and reliability. CONSTITUTION: A head related transfer function database(13) gathers a head related transfer function in a database when a speaker(1-7) and a listener(30) are located within a standard radius by channel. A location information unit(21) detects a location information with an angle, a distance between the listener(30) and the multichannel speaker(23) when a sound source(11) is inputted. A spatial information unit(12) removes a cross-talk phenomenon of the speakers(23) positioned within a preset area by channel by using the head related transfer function database(13) and the stored spatial information. A signal distributer(22) distributes signals to the speakers after the location information unit(21) obtains the speaker information when the sound source(11) is transferred to the signal distributer(22) through a cross-talk remover(14).
摘要翻译：目的：提供一种通过去除多通道扬声器中的串扰来实时再现3D声音的方法和系统，以改善声音位置效果，增强环境效果，实现优化的3D声音并显着提高产品质量和可靠性。构成：当讲话者（1-7）和收听者（30）位于标准半径内时，头部相关传递函数数据库（13）在数据库中收集头部相关传递函数。当声源（11）被输入时，位置信息单元（21）检测具有角度的位置信息，聆听者（30）与多声道扬声器（23）之间的距离。空间信息单元（12）通过使用头部相关传递函数数据库（13）和存储的空间信息，去除位于预定区域内的扬声器（23）的串扰现象。当声源（11）通过串扰消除器（14）传送到信号分配器（22）时，信号分配器（22）在位置信息单元（21）获得扬声器信息之后向扬声器分配信号。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式