会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 3. 发明公开
    • 카메라용 유무선 음성제어 릴리즈셔터
    • 有线/无线语音控制相机发行快门
    • KR1020070011765A
    • 2007-01-25
    • KR1020050066310
    • 2005-07-21
    • 김귀철
    • 김귀철
    • H04N5/232
    • H04N5/232G03B9/08G10L17/06
    • A cable/wireless voice-controlled release shutter of a camera is provided to easily take a picture of an object by converting a voice using a shutter and a halt-shutter of the camera and recording shutter and half-shutter voice signals. A cable/wireless voice-controlled release shutter of a camera includes a wireless transmitting unit(1) for generating a voice signal and a voice processing receiver(2) directly connected to the camera. The wireless transmitting unit includes a microphone(4) receiving the voice of a user and an RF transmitter(11) for transmitting the voice signal to the voice processing receiver. The voice processing receiver includes a sensor(16) for sensing pushing of a recording button(10), and a voice recorder(15) for recording shutter and half-shutter voice signals. The voice processing receiver further includes a voice converter(14) for recognizing a voice inputted through a cable microphone(5) or an RF receiver(12) and converting the voice into a recognizable signal, an MPU(Micro Processing Unit)(13) for recognizing the voice signal as a shutter or half-shutter signal, and a coupler(18) for dropping down a shutter voltage to the shutter and half-shutter signals.
    • 提供相机的有线/无线语音控制释放快门以通过使用快门和相机的停止快门转换语音以及记录快门和半快门语音信号来容易地拍摄对象的图片。 相机的有线/无线语音控制释放快门包括用于生成语音信号的无线发送单元(1)和直接连接到相机的语音处理接收器(2)。 无线发射单元包括接收用户语音的麦克风(4)和用于将语音信号发送到语音处理接收机的RF发射机(11)。 语音处理接收机包括用于感测按压记录按钮(10)的传感器(16)和用于记录快门和半快门语音信号的语音记录器(15)。 语音处理接收机还包括用于识别通过电缆麦克风(5)或RF接收机(12)输入的语音并将语音转换为可识别信号的语音转换器(14),MPU(微处理单元)(13) 用于将声音信号识别为快门或半快门信号,以及用于将快门电压降低到快门和半快门信号的耦合器(18)。
    • 5. 发明公开
    • 개별화된 핫워드 검출 모델들
    • 个性化的热点检测模型
    • KR20180010923A
    • 2018-01-31
    • KR20160099402
    • 2016-08-04
    • G10L15/22G10L13/08G10L15/04G10L17/04
    • G10L17/04G10L15/02G10L15/063G10L15/07G10L15/075G10L15/1815G10L17/06G10L17/08G10L17/18G10L17/24G10L2015/0638G10L2015/088
    • 시스템내에알림들을표시하기위한, 컴퓨터저장매체상에서인코딩된컴퓨터프로그램들을포함하는, 방법들, 시스템들및 장치들이개시된다. 일양태에서, 방법은사용자에의해말해진등록발언을나타내는등록음향데이터를획득하는단계; 다른사용자들에의해말해진발언들을나타내는후보음향데이터의세트를획득하는단계; 후보음향데이터의세트의각 후보음향데이터에대하여, 등록음향데이터와후보음향데이터의유사성을나타내는유사성스코어를결정하는단계; 유사성스코어들에적어도기초하여후보음향데이터의세트로부터후보음향데이터의서브세트를결정하는단계; 후보음향데이터의서브세트에기초하여검출모델을생성하는단계; 그리고사용자에의해말해진발언을검출하기위해상기검출모델을제공하는단계의동작들을포함한다.
    • 包括编码在计算机存储介质上的计算机程序的方法,系统和装置,用于在系统中呈现通知。 在一个方面,一种方法包括以下动作:获取表示用户说出的注册话语的注册声学数据,获得表示由其他用户讲话的话语的一组候选声学数据,针对该组候选声学数据的每个候选声学数据 ,表示登记声学数据与候选声学数据之间的相似性的相似性分数,至少基于相似性分数从该组候选声学数据中选择候选声学数据的子集,基于子集 候选声学数据,并提供用于检测用户说出的话语的检测模型。
    • 7. 发明授权
    • 스마트폰 기반 음성인식 어플리케이션 제어 방법
    • KR101779358B1
    • 2017-09-18
    • KR1020160161521
    • 2016-11-30
    • 동서대학교산학협력단주식회사 제이컴즈
    • 황기현김정아
    • G10L17/04G10L17/06G10L15/26G10L15/30
    • G10L17/04G06F21/31G10L15/26G10L15/30G10L17/06
    • 본발명의스마트폰기반음성인식어플리케이션제어방법은, 스마트폰에서실행된어플리케이션에서아이디및 비밀번호를네트워크로보내어서버프로그램으로부터사용자인증받는인증단계; 상기인증된사용자인경우, 어플리케이션에의해제어부가서버프로그램으로부터다수개의텍스트및 음성데이터를수신하는데이터수신단계; 상기제어부가다수개의텍스트및 음성데이터를통해각각의발음평가를할 수있도록화면을구성하는화면구성단계; 각각의발음평가진행시스마트폰의음성입력부를통해사용자음성신호를입력받은후, 상기음성입력부를통해입력된각각의발음평가에대한사용자음성신호를제어부에서사용자음성데이터로변환하는음성데이터변환단계; 상기제어부에서변환된사용자음성데이터와서버프로그램으로부터수신한음성데이터를비교한후, 발음정확도, 말의속도, 억양유사도, 또는목소리의크기에대한결과데이터를도출하는결과도출단계; 상기제어부에서도출된결과데이터를통해다수개의발음평가에대한결과를발음진단결과화면에나타내는결과출력단계; 를포함하는것으로, 본발명스마트폰기반음성인식어플리케이션제어방법은스마트폰에서입력받은사용자음성을서버프로그램에서수신된아나운서등의정확한표준발음과비교하여발음정확도, 말의속도, 목소리크기, 억양유사도와얼마나유사한지발음분석알고리즘을통해분석하고사용자가발음하기힘든유형을분류해발음연습을효과적으로할 수있게도와주는현저한효과가있다.
    • 10. 发明授权
    • 사용자 인식 장치 및 그에 의한 사용자 인식 방법
    • 用于检测用户的装置和用于检测用户的方法
    • KR100833096B1
    • 2008-05-29
    • KR1020070005798
    • 2007-01-18
    • 한국과학기술연구원
    • 권순일
    • B25J13/00B25J9/00
    • B25J13/003G10L17/06
    • A user detecting apparatus and a user detecting method by the same are provided to recognize a user with a short voice and with high speed operation by sorting characteristic vectors of a voice signal of the user and comparing the vectors with a user model. A user detecting apparatus comprises a characteristic vector sorting part, and a judging part(40). The characteristic vector sorting part compares a real voice signal with a user model and divides the real voice signal into a first characteristic vector without a user detecting error, and a second characteristic vector causing a user detecting error. The judging part judges a user only with the first characteristic vector. The user detecting apparatus is applied to a home service robot. The home service robot includes an input part(10), a characteristic vector abstract part(20), a characteristic vector sorting part(30), the judging part, and a control part(50). The control part controls a command operating part(60) according to the signal of the judging part.
    • 提供了一种用户检测装置和用户检测方法,用于通过对用户的语音信号的特征向量进行排序并将向量与用户模型进行比较来识别具有短语音和高速操作的用户。 用户检测装置包括特征向量分类部分和判断部分(40)。 特征矢量分类部分将真实语音信号与用户模型进行比较,并将真实语音信号划分成不具有用户检测误差的第一特征矢量,以及引起用户检测错误的第二特征向量。 判断部仅使用第一特征向量来判定用户。 用户检测装置被应用于家庭服务机器人。 家庭服务机器人包括输入部分(10),特征向量抽象部分(20),特征向量分类部分(30),判断部分和控制部分(50)。 控制部根据判断部的信号来控制指令运转部(60)。