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
    • 음성 신호 부호화 방법 및 장치
    • 编码语音信号的方法和装置
    • KR1020120109539A
    • 2012-10-08
    • KR1020127017163
    • 2010-12-10
    • 엘지전자 주식회사연세대학교 산학협력단
    • 전혜정김대환정규혁이민기강홍구이병석김락용
    • G10L19/04
    • G10L19/07G10L2019/0007G10L2019/001G10L2019/0013G10L2019/0016G10L19/06G10L19/032G10L19/04G10L19/08G10L19/09G10L19/107
    • PURPOSE: A voice signal encoding method and an apparatus thereof are provided to increase voice quality of a synthesized voice signal and to seek an optimal parameter which minimizes quantization error. CONSTITUTION: A voice signal encoding apparatus obtains a linear prediction filter coefficient of a current frame from an input signal by using linear prediction(S400). The apparatus obtains a quantized spectrum candidate vector of the current frame corresponding to the linear prediction filter coefficient based on first best information(S410). The apparatus interpolates the quantized spectrum candidate vector of the current frame and the quantized spectrum candidate vector of the previous frame(S420). [Reference numerals] (AA) Start; (BB) End; (S400) Obtaining linear prediction filter coefficient through linear prediction about an input signal; (S410) Obtaining a quantized spectrum candidate vector corresponding to the linear prediction filter coefficient by using first best information; (S420) Interpolating the quantized spectrum candidate vector; (S430) Generating a weighted value input signal by using a weighted filter for which the quantized spectrum candidate vector is used; (S440) Obtaining an adaptive code book of the weighted filter input signal; (S450) Searching for a fixed code book of a target signal of the fixed code book; (S460) Generating the signal by using the adaptive code book and the fixed code book; (S470) Generating a synthesis signal by using the signal and the synthesis filter; (S480) Obtaining a coding parameter which minimizes an error between an input signal and the synthesis
    • 目的:提供一种语音信号编码方法及其装置,以增加合成语音信号的语音质量,并寻找最小化量化误差的最优参数。 构成:语音信号编码装置通过使用线性预测从输入信号获得当前帧的线性预测滤波器系数(S400)。 该装置基于第一最佳信息获得对应于线性预测滤波器系数的当前帧的量化频谱候选向量(S410)。 该装置内插当前帧的量化频谱候选矢量和前一帧的量化频谱候选矢量(S420)。 (附图标记)(AA)开始; (BB)结束; (S400)通过关于输入信号的线性预测获得线性预测滤波器系数; (S410)通过使用第一最佳信息获得与线性预测滤波器系数对应的量化频谱候选矢量; (S420)插值量化频谱候选矢量; (S430)使用使用了量化频谱候补矢量的加权滤波器来生成加权值输入信号; (S440)获取加权滤波器输入信号的自适应码本; (S450)搜索固定码本的目标信号的固定码本; (S460)使用自适应码本和固定码本生成信号; (S470)使用信号和合成滤波器生成合成信号; (S480)获得最小化输入信号与合成之间的误差的编码参数
    • 4. 发明公开
    • 오디오 신호 처리 방법 및 장치
    • 音频信号处理方法和装置
    • KR1020120098701A
    • 2012-09-05
    • KR1020127012638
    • 2010-11-24
    • 엘지전자 주식회사연세대학교 산학협력단
    • 전혜정김대환강홍구이민기이병석정규혁
    • G10L19/00G11B20/10
    • G10L19/00G10L19/005G10L19/12G10L2019/0001G10L19/028G10L19/04G10L19/08G10L19/125G11B20/10
    • PURPOSE: An audio signal processing method and a device thereof are provided to minimize propagation of errors to a next frame. CONSTITUTION: An audio signal is received wherein the audio signal includes current frame(S100). An error occurs in data of the current frame(S300). Frame error concealment about data of the current frame is performed. A first provisional output signal of the current frame is generated(S400). A parameter is generated by performing at least one of short term prediction, long term prediction, and fixed codebook search(S500). [Reference numerals] (S100) Receiving an audio signal including current frame data; (S200) Checking generation of an error; (S300) Error has occurred?; (S400) Generating a temporary output signal by performing error concealment; (S500) Re-encoding the temporary output signal; (S600) Transferring encoded parameters to a decoding unit; (S700) Decoding the current frame
    • 目的:提供一种音频信号处理方法及其装置,以最小化误差到下一帧的传播。 构成:接收音频信号,其中音频信号包括当前帧(S100)。 在当前帧的数据中发生错误(S300)。 执行关于当前帧的数据的帧错误隐藏。 产生当前帧的第一临时输出信号(S400)。 通过执行短期预测,长期预测和固定码本搜索中的至少一个来生成参数(S500)。 (附图标记)(S100)接收包括当前帧数据的音频信号; (S200)检查错误的生成; (S300)发生错误? (S400)通过执行错误隐藏生成临时输出信号; (S500)重新编码临时输出信号; (S600)将编码参数传送到解码单元; (S700)解码当前帧
    • 8. 发明授权
    • 오디오 신호 처리 방법 및 장치
    • 用于处理音频信号的方法和设备
    • KR101804922B1
    • 2017-12-05
    • KR1020127026235
    • 2011-03-23
    • 엘지전자 주식회사
    • 정규혁김대환이창헌김락용전혜정이병석강인규
    • G10L19/12
    • G10L19/04G10L19/06G10L19/22G10L19/24
    • 본발명은오디오신호의현재프레임에대해서스펙트럼분석을수행함으로써, 상기현재프레임이제 1 대역및 제 2 대역을복수의대역들중 어디에해당하는지를나타내는대역폭정보를결정하는단계; 상기대역폭정보를근거로, 현재프레임에대응하는차수정보를결정하는단계; 상기현재프레임에대해선형-예측분석을수행함으로써, 제 1 차수의제 1 셋선형-예측변환계수를생성하는단계; 상기제 1 셋선형-예측변환계수를벡터-양자화함으로써, 제 1 셋인덱스를생성하는단계; 상기현재프레임에대해선형-예측분석을수행함으로써, 상기차수정보에따라서, 제 2 차수의제 2 셋선형-예측변환계수를생성하는단계; 및, 상기제 2 셋선형-예측변환계수가생성된경우, 상기제 1 셋인덱스및 상기제 2 셋선형-예측변환계수를이용하여제 2 셋차분에대해벡터-양자화를수행하는단계를포함하는오디오신호처리방법을개시한다.
    • 本发明提供了一种方法,包括:通过对音频信号的当前帧执行频谱分析,确定指示当前帧现在对应于第一频带和第二频带的多个频带中的哪一个频带的带宽信息; 基于带宽信息确定与当前帧对应的度信息; 对当前帧执行线性预测分析以产生第一组第一组线性预测变换系数; 通过矢量量化第一组线性预测变换系数产生第一组索引; 通过对当前帧执行线性预测分析,根据顺序信息生成第二组第二组线性预测变换系数; 并且当产生第二组线性预测系数时,使用第一组索引和第二组线性预测变换系数对第二组差异执行矢量量化, 公开了一种音频信号处理方法。