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
    • 72. 发明公开
    • 오디오 신호 인코더
    • 音频信号编码器
    • KR1020160099684A
    • 2016-08-22
    • KR1020167019246
    • 2013-12-17
    • 노키아 테크놀로지스 오와이
    • 바실라체아드리아나라모안씨사카리라크소넨라세주하니
    • G10L19/07G10L19/038G10L19/008
    • G10L19/038G10L19/07G10L19/008
    • 장치는적어도하나의오디오신호를규정하는파라미터의적어도하나의벡터를발생하도록구성된벡터발생기; 파라미터의연계된적어도하나의순서화된벡터를발생하기위해적어도하나의벡터절대튜플의순서화에따라파라미터의적어도하나의벡터를분류하도록구성된격자벡터양자화기를포함하고, 격자벡터양자화기는리더클래스의리스트로부터적어도하나의잠재적인코드벡터를선택하도록구성되고; 격자벡터양자화기는적어도하나의잠재적인코드벡터와파라미터의적어도하나의순서화된벡터사이의거리를결정하도록구성되고; 격자벡터양자화기는최소연계된거리를발생하는잠재적인코드벡터와연계된적어도하나의리더클래스를결정하도록구성되고; 격자벡터양자화기는출력격자양자화된코드벡터를발생하기위해적어도하나의리더클래스를전치하도록구성된다.
    • 该设备包括:矢量生成器,被配置为生成定义至少一个音频信号的至少一个参数矢量; 以及格型矢量量化器,其被配置为根据至少一个矢量绝对元组的排序对至少一个参数向量进行分类以生成与所述参数相关联的至少一个有序向量, 配置为选择至少一个潜在的代码向量; 其中格型向量量化器被配置为确定至少一个潜在码向量的至少一个有序向量与参数之间的距离; 其中格型矢量量化器被配置为确定与生成最小相关距离的潜在代码矢量相关联的至少一个阅读器类别; 格矢量量化器被配置为转置至少一个领导者类以生成输出格量化代码向量。
    • 74. 发明公开
    • 벡터 양자화 장치, 벡터 역양자화 장치, 및 이러한 방법
    • 矢量量化器,矢量反相量子和方法
    • KR1020100085908A
    • 2010-07-29
    • KR1020107007679
    • 2008-10-10
    • 파나소닉 주식회사
    • 사토가오루모리도시유키에하라히로유키
    • G10L19/12G10L19/07G10L19/038
    • G10L19/032G10L19/07G10L19/18G10L2019/0005
    • A vector quantizer which improves the accuracy of vector quantization in switching over a vector quantization codebook on a first stage depending on the type of feature having the correlation with a quantization target vector. In the vector quantizer, a classifier (101) generates classification information representing a type of narrowband LSP vector having the correlation with wideband LSP (Line Spectral Pairs) out of the plural types. A first codebook (103) selects one sub-codebook corresponding to the classification information as a codebook used for the quantization of the first stage from plural sub-codebooks (CBa1 to CBan) corresponding to each of the types of narrowband LSP vectors. A multiplier (107) multiplies the quantization residual vector of the first stage inputted from an adder (104) by a scaling factor corresponding to the classification information out of plural scaling factors stored in a scaling factor determining section (106) and outputs it to an adder (109) as the quantization target of a second stage.
    • 一种矢量量化器,其根据具有与量化目标矢量相关的特征的类型,在第一级切换矢量量化码本时提高矢量量化的精度。 在矢量量化器中,分类器(101)生成表示与多种类型中的宽带LSP(线谱对)相关的窄带LSP矢量的类型的分类信息。 第一码本(103)从与各种窄带LSP矢量对应的多个子码本(CBa1〜CBan)中选择与分类信息对应的一个子码本作为用于第一级量化的码本。 乘法器(107)将从加法器(104)输入的第一级的量化残差矢量与存储在比例因子确定部分(106)中的多个缩放因子中的分类信息相对应的缩放因子相乘,并将其输出到 加法器(109)作为第二级的量化目标。
    • 77. 发明公开
    • 광대역 음성 부호화를 위한 엘에스에프 계수 벡터 양자화기
    • 用于宽带语音编码的线频率系数矢量量测器
    • KR1020040078760A
    • 2004-09-13
    • KR1020030013606
    • 2003-03-05
    • 한국전자통신연구원
    • 성호상황대환강상원이강은
    • G10L19/07G10L19/038
    • G10L19/07
    • PURPOSE: An LSF(Line Spectral Frequency) coefficient vector quantizer for broadband voice coding is provided to reduce memory capacity and the quantity of calculations required for quantization and prevent the deterioration of the performance of the quantizer. CONSTITUTION: An LSF coefficient vector quantizer includes a prediction quantizer(30), an non-prediction quantizer(31), and a switch(32). The prediction quantizer includes the first vector quantizer(VQ1) for non-structurally quantizing an LSF coefficient vector to produce a candidate vector to be quantized, a predictor for calculating a predicted LSF vector of the LSF coefficient vector, and the first lattice quantizer for lattice-quantizing the candidate vector with reference to the predicted LSF vector to produce a final prediction quantization vector of the LSF coefficient vector. The non-prediction quantizer includes the second vector quantizer(VQ2) for non-structurally quantizing the LSF coefficient vector to produce a candidate vector to be quantized, and the second lattice quantizer for lattice-quantizing the candidate vector to produce a final non-prediction quantization vector of the LSF coefficient vector. The switch decides one of the final prediction quantization vector and the final non-prediction quantization vector, which has smaller difference from the LSF coefficient vector as a final quantization vector of the LSF coefficient vector.
    • 目的:提供用于宽带语音编码的LSF(线谱频率)系数矢量量化器,以减少存储容量和量化所需的计算量,并防止量化器性能的恶化。 构成:LSF系数矢量量化器包括预测量化器(30),非预测量化器(31)和开关(32)。 预测量化器包括用于非结构量化LSF系数向量以产生要量化的候选矢量的第一矢量量化器(VQ1),用于计算LSF系数矢量的预测LSF矢量的预测器和用于晶格的第一晶格量化器 - 参考预测的LSF向量量化候选向量,以产生LSF系数向量的最终预测量化矢量。 非预测量化器包括用于非结构量化LSF系数向量以产生要量化的候选向量的第二矢量量化器(VQ2),以及用于对候选矢量进行晶格量化以产生最终非预测的第二格子量化器 LSF系数向量的量化向量。 开关决定作为LSF系数向量的最终量化矢量的与LSF系数矢量的差较小的最终预测量化矢量和最终的非预测量化矢量之一。
    • 78. 发明授权
    • 음성 인식 방법
    • 语音识别方法
    • KR100269357B1
    • 2000-10-16
    • KR1019980015696
    • 1998-04-30
    • 주식회사 엘지이아이
    • 이윤근김기백이병수이종석
    • G10L19/07G10L25/24
    • PURPOSE: A method for recognizing a voice is provided to improve an error generated when an LSP(Line Spectral Pairs) is used as a feature vector. CONSTITUTION: A voice is input. An LSP is extracted from the input voice. The extracted LSP is converted to a pseudo-cepstrum. The converted pseudo-cepstrum is used as a feature vector in a voice recognizing process. At this time, the word, cepstrum is formed by changing inversely an order of the front 'spec' of a spectrum. In the method, a predetermined mathematical expression is used for performing a converting process of the extracted LSP to the pseudo-cepstrum.
    • 目的:提供一种用于识别语音的方法,以改善当使用LSP(线谱对)作为特征向量时产生的错误。 声明:输入声音。 从输入语音中提取LSP。 提取的LSP被转换为伪倒谱。 转换的伪倒频谱被用作语音识别过程中的特征向量。 此时,倒频谱是通过反向改变频谱前面的“spec”的顺序来形成的。 在该方法中,使用预定的数学表达式来执行所提取的LSP到伪倒频谱的转换处理。