-
-
公开(公告)号：KR100903110B1
公开(公告)日：2009-06-16
申请号：KR1020070036585
申请日：2007-04-13
Applicant: 한국전자통신연구원 , 한양대학교 산학협력단 , 광운대학교 산학협력단 , 연세대학교 산학협력단
Abstract: 본 발명은 트렐리스 부호 양자화 알고리듬을 이용한 광대역 음성 부호화기용 LSF 계수 양자화 장치 및 방법에 관한 것으로, LSF 계수 벡터를 양자화하여 양자화 대상 후보 벡터를 산출하고, 상기 후보 벡터를 상기 LSF 계수 벡터의 예측된 LSF 벡터를 참조하여 트렐리스 부호 양자화(TCQ)하여 상기 LSF 계수 벡터의 예측 양자화 최종 벡터를 산출하는 예측 구조 양자화부, 상기 LSF 계수 벡터를 양자화하여 양자화 대상 후보 벡터를 산출하고, 상기 후보 벡터를 트렐리스 부호 양자화(TCQ)하여 상기 LSF 계수 벡터의 비예측 양자화 최종 벡터를 산출하는 비예측 구조 양자화부 및 상기 예측 양자화 최종 벡터와 비예측 양자화 최종 벡터 중 상기 LSF 계수 벡터와의 차가 작은 것을 상기 LSF 계수 벡터의 최종 양자화 벡터로 결정하는 스위칭부로 구성되어, 안전한 양자화와 에러 전달 현상을 줄여 할당되는 비트와 SD 성능를 개선시킬 수 있는 효율적인 LSF 양자화기를 제공한다.
양자화기, BCTCQ 알고리즘, 예측 벡터, 비예측 벡터-
公开(公告)号：KR1020080102027A
公开(公告)日：2008-11-24
申请号：KR1020070048302
申请日：2007-05-17
Applicant: 한국전자통신연구원 , 한양대학교 산학협력단 , 광운대학교 산학협력단 , 연세대학교 산학협력단
CPC classification number: H03M7/40 , G10L19/0017
Abstract: A lossless coding/decoding apparatus and method of audio signal are provided to improve the performance of an audio encoder by compressing audio signal into a bit string having a less bit number at the same coding distortion. A lossless encoding apparatus comprises a second encoder module(210) including a first encoder(230) directly encoding a first symbol and a plurality of second encoders(220) converting the first symbol into a second symbol and encoding the second symbol; a first selector(250) comparing the performance of the second encoders and the first encoder and outputting an encoding mode according to the comparison result; and a second selector(260) encoding the first symbol corresponding to the encoding mode and outputting the final bit string.
Abstract translation: 提供音频信号的无损编码/解码装置和方法，以通过将音频信号压缩成具有相同编码失真的位数较少的位串来提高音频编码器的性能。 无损编码装置包括第二编码器模块（210），其包括直接编码第一符号的第一编码器（230）和将第一符号转换为第二符号并对第二符号进行编码的多个第二编码器（220） 比较第二编码器和第一编码器的性能并根据比较结果输出编码模式的第一选择器（250） 以及编码对应于编码模式的第一符号并输出最终位串的第二选择器（260）。
-
公开(公告)号：KR1020040078760A
公开(公告)日：2004-09-13
申请号：KR1020030013606
申请日：2003-03-05
Applicant: 한국전자통신연구원
IPC: G10L19/07 , G10L19/038
CPC classification number: G10L19/07
Abstract: 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.
Abstract translation: 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。
-
公开(公告)号：KR100324204B1
公开(公告)日：2002-02-16
申请号：KR1019990062269
申请日：1999-12-24
Applicant: 한국전자통신연구원
IPC: H03M7/50
CPC classification number: G10L19/06
Abstract: 본발명은선형스펙트럼쌍(LSP : Line Spectrum Pair) 계수의예측분할행렬양자화(PSMQ : Predictive Split Metrix Quantization) 또는예측분할벡터양자화(PSVQ : Predictive Split Vector Quantization) 방식을사용하는음성부호화기에관한것으로서, 보다상세하게설명하면코드북탐색시요구되는계산량을줄이기위한예측분할벡터양자화및 예측분할행렬양자화방식에의한선스펙트럼쌍양자화기의고속탐색방법에관한것이다. 본발명에따른예측분할벡터양자화및 예측분할행렬양자화방식에선스펙트럼쌍양자화기의고속탐색방법은, 순서성질을갖는목표벡터와에러코드북을구성하고, 순서성질을갖도록재구성된각 코드북은선정된하나의특정열의요소값을근거로전체코드북을내림차순으로정렬한다. 그리고, 각코드북내 정렬된특정열의요소값과목표벡터내 해당열전후의요소값들을순방향및 역방향비교를통해순서성질을만족하는대상코드벡터들의검색영역의시작점과끝점을구한다. 이렇게시작점과끝점이구해지면, 이탐색범위내에서만계산량이많은에러기준값(criterion) E을구하여최적의대상코드벡터를구한다.
-
Patent Agency Ranking
公开(公告)号：KR100263296B1
公开(公告)日：2000-08-01
申请号：KR1019970059520
申请日：1997-11-12
Applicant: 한국전자통신연구원
IPC: G10L15/02
Abstract: PURPOSE: A voice activity measuring method for a G.729 voice encoder is provided to determined optimal transmission rate for implementing each input frame for adaptive variable rate voice encoder. CONSTITUTION: The voice activity measuring method for the G.729 voice encoder includes following steps. At first, a voice frame energy value obtained during linear estimation coefficient calculation is compared with a few adaptive threshold values, the voice activity is calculated, and a sub-transmission rate corresponding to the voice activity is estimated. At the second step, numerous characteristic parameters obtained during analysis are combined to perform phonetic classification in order to verify the validity of the estimated sub-transmission rate. At the third step, variable hang-over algorithm is performed based on signal to noise ratios and the characteristic parameters so as to minimize the voice quality degeneration at the end of voice during encoding at the sub-transmission rate determined at the first and the second step. At last, a conversion to the transmission rate 1/2 is performed so as to prevent voice degeneration due to the capabilities of the voice encoders during conversion from transmission rate 1 to transmission rate 1/8 while encoding at the sub-transmission rate determined at the first through third steps.
Abstract translation: 目的：提供一种用于G.729语音编码器的语音活动测量方法，用于确定实现用于自适应可变速率语音编码器的每个输入帧的最佳传输速率。 构成：G.729语音编码器的语音活动测量方法包括以下步骤。 首先，将在线性估计系数计算中获得的语音帧能量值与若干自适应阈值进行比较，计算语音活动，并且估计与语音活动相对应的子传输速率。 在第二步，将分析期间获得的许多特征参数组合起来进行语音分类，以验证估计的子传输速率的有效性。 在第三步骤中，基于信噪比和特征参数执行可变挂机算法，以便在以第一和第二确定的子传输速率进行编码期间使话音结束时语音质量退化最小化 步。 最后，执行到传输速率1/2的转换，以便在以传输速率1转换到传输速率1/8之前防止由于语音编码器的能力引起的语音退化，同时以按照 第一到第三步。
-
-
公开(公告)号：KR1019970011528B1
公开(公告)日：1997-07-11
申请号：KR1019930027344
申请日：1993-12-11
Applicant: 한국전자통신연구원
IPC: G02B27/10
Abstract: The optoisolator formed by isotropic etching and substrate binding and manufacturing method thereof comprising the first step isotropic etching a directional Si substrate(201) by KOH to form a pyramid type etching hole(200) inclined against the substrate bottom surface; the second step forming a optical reflection layer(202) on said etching hole(200) to enhance reflection rate and then forming a deposit layer(203) to flatten the surface for binding substrates; the third step grinding said deposit layer(203) and then binding a substrate(204) having the same direction as said substrate(201) on said deposit layer(203); and the forth step removing said substrate(201) to form a optical reflection surface(205) is disclosed. Thereby, it is possible to provide a small optoisolator which optic signal inputted during optic signal process can be divided.
Abstract translation: 通过各向同性蚀刻和衬底粘合形成的光隔离器及其制造方法包括：通过KOH对第一步骤各向同性蚀刻<410方向的Si衬底（201），形成倾斜于衬底底面的金字塔型蚀刻孔（200）; 第二步骤，在所述蚀刻孔（200）上形成光反射层（202），以提高反射率，然后形成沉积层（203）以平坦化用于粘结基片的表面; 第三步骤研磨所述沉积层（203），然后将具有与所述基底（201）相同方向的基底（204）结合在所述沉积层（203）上; 并且公开了去除所述衬底（201）以形成光学反射表面（205）的第四步骤。 由此，可以提供在光信号处理期间输入的光信号可以被分割的小型光隔离器。
-
公开(公告)号：KR1019970009972B1
公开(公告)日：1997-06-19
申请号：KR1019930028480
申请日：1993-12-18
Applicant: 한국전자통신연구원
IPC: H01L21/30
Abstract: The present invention provides a method of making an ink jet nozzle by using anisotropic etching and hetero junction. This method includes the steps of joining a substrate (101) with a crystal direction (100) and a substrate (102) with a different crystal direction (101), and making the substrate (102) thin to a given thickness; forming an etching protecting mask on the substrate (101), selectively etching the same by anisotropic etching to form a region (105) in an inverted pyramid shape, and exposing the substrate (102)'s given portion to form an etching protecting window (107); and anisotropic etching the substrate (102) through the opened window (107) to form a removed region (108) of rectangular shape.
Abstract translation: 本发明提供一种通过使用各向异性蚀刻和异质结制造喷墨嘴的方法。 该方法包括以下步骤：使晶体方向（100）的衬底（101）与具有不同晶体方向（101）的衬底（102）接合，并使衬底（102）变薄至给定厚度; 在所述基板（101）上形成蚀刻保护掩模，通过各向异性蚀刻选择性地蚀刻所述蚀刻保护掩模，以形成倒金字塔形状的区域（105），并暴露所述基板（102）的给定部分以形成蚀刻保护窗口 107）; 并通过打开的窗口（107）对衬底（102）进行各向异性蚀刻，以形成矩形形状的去除区域（108）。
-
-
-
-
-
-
-
-
-
Search Results
153
records
(took 0.02 seconds)
