-
公开(公告)号：US5007101A
公开(公告)日：1991-04-09
申请号：US454022
申请日：1982-12-28
摘要： An auto-correlation unit for pattern recognition to obtain auto-correlation functions as for sampled signals is disclosed N pieces of sample values X.sub.n (n=0 to N-1)are extracted from a series of the sample values expressed with an accuracy of optional multi-bits and the auto-correlation coefficients of these N pieces of the sample values obtained. The sum of sample values X.sub.n and X.sub.n-.tau. (.tau.=0 to P) are calculated. The squared value (X.sub.n +X.sub.n-.tau.).sup.2 of the added results is previously memorized in a ROM while being addressed by the resultant sum. The auto-correlation coefficients are available by feeding the output (X.sub.n +X.sub.n-.tau.).sup.2 of the ROM and executing calculation as defined by the following equation: ##EQU1##
摘要翻译： 公开了一种用于模式识别以获得与采样信号相关的自相关函数的自相关单元，从采用可选择的精度表示的一系列采样值中提取出N个采样值Xn（n = 0至N-1） 这些N个样本值获得的多位和自相关系数。 计算采样值Xn和Xn-tau（τ= 0至P）的总和。 相加结果的平方值（Xn + Xn-τ）2预先存储在ROM中，同时通过所得到的和来寻址。 通过馈送ROM的输出（Xn + Xn-τ）2并执行由以下等式定义的计算，可以获得自相关系数：
-
公开(公告)号：US4596031A
公开(公告)日：1986-06-17
申请号：US451937
申请日：1982-12-21
CPC分类号： G10L15/00
摘要： In a speech recognition system, similarity calculations between speech feature patterns are reduced by stopping similarity calculations for any one reference pattern when a frame in the pattern fails to exceed a corresponding similarity threshold.
摘要翻译： 在语音识别系统中，当模式中的帧不能超过对应的相似性阈值时，通过停止任何一个参考模式的相似度计算来减少语音特征模式之间的相似度计算。
-
公开(公告)号：US4633499A
公开(公告)日：1986-12-30
申请号：US433522
申请日：1982-10-08
摘要： In a speech analysis/recognition system, the input utterance is also directly registered (stored) for later readout verification of correct start and end control signals.
摘要翻译： 在语音分析/识别系统中，输入语音也被直接注册（存储），用于后续读出验证正确的开始和结束控制信号。
-
公开(公告)号：US4516215A
公开(公告)日：1985-05-07
申请号：US416165
申请日：1982-09-09
CPC分类号： G10L15/00
摘要： Speech recognition using self-correlation (auto-correlation) parameters and time normalization is improved by first making a preliminary decision using lower-order (zeroth to third) self-correlation coefficients derived from the self-correlation function of order m: ##EQU1## where .phi.(0) is the power, .phi.(1) is the linear or first order coefficient, and so on.
摘要翻译： 使用自相关（自相关）参数和时间归一化的语音识别通过首先使用从阶数m的自相关函数导出的低阶（第零至第三）自相关系数进行初步判定得到改善：
其中phi（0）是幂，phi（1）是线性或一阶系数，依此类推。 -
公开(公告)号：US4618983A
公开(公告)日：1986-10-21
申请号：US452298
申请日：1982-12-22
CPC分类号： G10L15/12
摘要： Speech recognition executes the final matching operation for each characteristic pattern according to the preliminarily selected order prior to the final matching, where any further calculation in reference to such a characteristic pattern is not executed after a specific condition has been satisfied during the matching calculation, and as result, calculation requirements can be drastically minimized so that the speech recognition can eventually be implemented at higher speed and yet under a sharply reduced cost.
摘要翻译： 语音识别根据最终匹配之前的预先选择的顺序执行每个特征图案的最终匹配操作，其中在匹配计算中已经满足特定条件之后，不执行参考这种特征图案的任何进一步计算，以及 因此，计算要求可以大大降低，从而语音识别最终可以以更高的速度实现，而且大大降低成本。
-
公开(公告)号：US4908864A
公开(公告)日：1990-03-13
申请号：US034060
申请日：1987-04-02
IPC分类号： G10L15/06
CPC分类号： G10L15/063
摘要： Inputted voice signals are analyzed in units of syllables. Each syllable is compared with standard syllables preregistered in a memory and thereby recognized but corrections may be made on erroneous recognitions by referencing a dictionary or entering a command. Each standard pattern is associated with phonological information on the neighborhood in a voice signal from where it was extracted and may be updated by another pattern having the same phonological information. Temporal sequences of correct and erroneous recognitions of individual syllables as well as whole syllables are stored and referenced in determining whether a standard pattern should be updated. A maximum pattern number may be set for each syllable.
摘要翻译： 输入的语音信号以音节为单位进行分析。 将每个音节与预先登记在存储器中的标准音节进行比较，从而识别，但可以通过引用字典或输入命令来对错误的识别进行修正。 每个标准模式与来自其提取的语音信号中的邻域的语音信息相关联，并且可以由具有相同语音信息的另一模式更新。 在确定是否应更新标准模式时，存储和引用单个音节以及整个音节的正确和错误识别的时间序列。 可以为每个音节设置最大花样编号。
-
公开(公告)号：US4775951A
公开(公告)日：1988-10-04
申请号：US935095
申请日：1986-11-24
摘要： A correlation function computing device capable of computing a selfcorrelation function or the like is constructed to comprise a window function memorizing circuit, an A/D converting circuit for converting time series of a voice signal into digital values, a multiplier for calculating products of the window functions and the digitalized time series of the voice signal and also products required for the computation of the correlation function, a memory circuit for storing the products of the window functions and the digitalized time series of the voice signal delivered from the multiplier, a selector circuit for transferring data supplied to the multiplier, an adder for obtaining a product-sum in a divided manner, a memory circuit of a first-in and first-out type for storing the product-sum dividedly obtained in the adder, and a control circuit for controlling all of the aforementioned circuits.
摘要翻译： 能够计算自相关函数等的相关函数计算装置被构造成包括窗口函数存储电路，用于将语音信号的时间序列转换为数字值的A / D转换电路，用于计算窗口的乘积的乘法器 功能和语音信号的数字化时间序列，以及计算相关功能所需的产品，存储电路，用于存储窗口功能的乘积和从乘法器传送的语音信号的数字化时间序列，选择器电路 用于传送提供给乘法器的数据，用于以分割方式获得乘积和的加法器，用于存储在加法器中分割获得的积和的先入先出类型的存储电路和控制电路 用于控制所有上述电路。
-
8.发明授权公开(公告)号：US06239788B1
公开(公告)日：2001-05-29
申请号：US09131394
申请日：1998-08-07
IPC分类号： G09G500
CPC分类号： G06F3/0412 , G02F1/13338 , G06F3/044 , G09G3/3611
摘要： In a coordinate detection period, a first electrode drive circuit sequentially selects source electrodes Sn of the TFT LCD panel to connect the same to an x-signal current amplifying circuit, and to apply an AC voltage to the non-selected electrodes. The x-signal current amplifying circuit detects a detection current flowing through the selected electrode, the magnitude of which corresponds to a distance to a finger (or a conductor pen), to obtain an x-coordinate detection voltage. A second electrode drive circuit and a y-signal current amplifying circuit operate in similar fashion to obtain a y-coordinate detection voltage. A coordinate detection circuit then obtains the x- and y-coordinate values of the designated position based on the x-coordinate detection voltage and the y-coordinate detection voltage, thus enabling pointing by a codeless pen and a finger.
摘要翻译： 在坐标检测期间，第一电极驱动电路顺序地选择TFT LCD面板的源电极Sn将其连接到x信号电流放大电路，并向未选择的电极施加AC电压。 x信号电流放大电路检测流过所选电极的检测电流，其大小对应于与手指（或导体笔）的距离，以获得x坐标检测电压。 第二电极驱动电路和y信号电流放大电路以类似的方式工作以获得y坐标检测电压。 然后，坐标检测电路基于x坐标检测电压和y坐标检测电压获得指定位置的x坐标值和y坐标值，从而能够通过无代码笔和手指指示。
-
公开(公告)号：US5953454A
公开(公告)日：1999-09-14
申请号：US657671
申请日：1996-05-31
申请人： Hiroyuki Iwahashi
发明人： Hiroyuki Iwahashi
摘要： A minimum distance storage device for storing distances obtained in pattern recognition as arranged in the ascending order easily generates addresses with no need for multiloop control and at the same time of detects a minimal distance of the same category. A read-out distance read from a result memory and temporarily held in a first register, and an input distance input from a distance calculating portion and temporarily held in a second register, are compared with each other by a second selector. The read-out distance, if smaller than the input distance, is written in the same place where it was stored in the result memory. If the read-out distance is larger than the input distance, the latter is written in the place where the read-out distance was stored and, then, the larger read-out distance and other read-out distances are written in places moved to the right each by one in sequence in the result memory.
摘要翻译： 用于存储以升序排列的模式识别中获得的距离的最小距离存储装置容易地生成地址，而不需要多循环控制，并且同时检测相同类别的最小距离。 从结果存储器读取并临时保存在第一寄存器中的读出距离和从距离计算部分输入的临时保持在第二寄存器中的输入距离通过第二选择器彼此进行比较。 读取距离（如果小于输入距离）写入存储在结果存储器中的相同位置。 如果读出距离大于输入距离，则将其写入存储读出距离的位置，然后将较大的读出距离和其他读出距离写入移动到 在结果存储器中依次排列右边的权限。
-
10.发明授权Phonemic classification in speech recognition system having accelerated response time 失效具有加速响应时间的语音识别系统中的音素分类
公开(公告)号：US4937869A
公开(公告)日：1990-06-26
申请号：US706528
申请日：1985-02-28
申请人： Hiroyuki Iwahashi , Toru Ueda
发明人： Hiroyuki Iwahashi , Toru Ueda
CPC分类号： G10L15/00
摘要： An acoustic processing unit for a speech recognition system is characterized in that the parameters are extracted from the speech data, including correlation functions, the zero crossing number of the original waveforms, the zero crossing number of the differential waveforms, and the average level of the waveforms. A suitable threshold is selected from a plurality of thresholds preliminarily stored, depending on the inputted speech volume level. The inputted speech volume level of the speaker is detected so that feedback of the detected volume level is obtained. The selected threshold set is then compared with each of the parameters to thereby make the phonemic classification. Because the plurality of threshold sets are thus automatically selected depending on the inputted speech volume level, flexible phonemic classification can be obtained to exactly detect the speech sections.
摘要翻译： 用于语音识别系统的声学处理单元的特征在于，从语音数据中提取参数，包括相关函数，原始波形的零交叉数，差分波形的零交叉数以及平均电平 波形。 根据输入的语音音量水平，从预先存储的多个阈值中选择合适的阈值。 检测扬声器的输入语音音量，从而获得检测音量级别的反馈。 然后将所选择的阈值集合与每个参数进行比较，从而进行音素分类。 因为根据输入的语音音量水平自动选择多个阈值集合，所以可以获得灵活的音素分类以准确地检测语音部分。
-
-
-
-
-
-
-
-
-
搜索结果
10 条记录 (耗时 0.019 秒)
