专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

81. 发明授权

US5276629A Method and apparatus for wave analysis and event recognition 失效
标题翻译：用于波分析和事件识别的方法和装置
公开(公告)号：US5276629A
公开(公告)日：1994-01-04
申请号：US930476
申请日：1992-08-14
申请人： Kentyn Reynolds
发明人： Kentyn Reynolds
IPC分类号： G01H3/06 , G01R23/00 , G01R23/16 , G01R23/18 , G06K9/00 , G10L11/02 , G10L15/02 , G10L15/10 , G06F15/31
CPC分类号： G10L15/02 , G06K9/00536 , G10L25/87
摘要： A method and apparatus for acquiring, recording, synchronizing, analyzing, and interpreting continuous wave data. A process for isolating separate events from wave data composed of multiple events and for determining the identification and characteristics of an event's wave source, its frequency components, amplitude, duration, and timing. A procedure for defining the parameters of the source wave and for adapting these parameters to the analysis requirements (e.g., correct frequency, amplitude, and timing divisions). A procedure for verifying interpretation results, for correcting interpretation errors, and for retaining successful results. Products of the processes, such as databases and recordings, can be made.
摘要翻译：一种用于获取，记录，同步，分析和解释连续波数据的方法和装置。用于将单独的事件与由多个事件组成的波形数据隔离并用于确定事件波源的识别和特性，其频率分量，振幅，持续时间和定时的过程。用于定义源波的参数并使这些参数适应分析要求（例如，正确的频率，幅度和时间分割）的过程。验证解释结果的过程，纠正解释错误以及保留成功的结果。可以制作流程的产品，如数据库和录音。

82. 发明授权

US4956865A Speech recognition 失效
标题翻译：语音识别
公开(公告)号：US4956865A
公开(公告)日：1990-09-11
申请号：US191824
申请日：1988-05-02
申请人： Matthew Lennig , Paul Mermelstein , Vishwa N. Gupta
发明人： Matthew Lennig , Paul Mermelstein , Vishwa N. Gupta
IPC分类号： G10L11/02 , G10L15/02
CPC分类号： G10L25/87 , G10L15/02
摘要： In a speech recognizer, for recognizing unknown utterances in isolated-word speech or continuous speech, improved recognition accuracy is obtained by augmenting the usual spectral representation of the unknown utterance with a dynamic component. A corresponding dynamic component is provided in the templates with which the spectral representation of the utterance is compared. In preferred embodiments, the representation is mel-based cepstral and the dynamic components comprise vector differences between pairs of primary cepstra. Preferably the time interval between each pair is about 50 milliseconds. It is also preferable to compute a dynamic perceptual loudness component along with the dynamic parameters.
摘要翻译：在语音识别器中，为了识别孤立词语音或连续语音中的未知语音，通过用动态分量增加未知语音的常规频谱表示来获得改进的识别精度。在模板中提供相应的动态分量，与之对比发音的频谱表示。在优选实施例中，该表示是基于mel的倒频谱，并且动态分量包括主要cepstra对之间的矢量差异。优选地，每对之间的时间间隔约为50毫秒。还优选地计算动态感知响度分量以及动态参数。

83. 发明授权

US4833714A Speech recognition apparatus 失效
标题翻译：语音识别装置
公开(公告)号：US4833714A
公开(公告)日：1989-05-23
申请号：US228149
申请日：1988-08-04
申请人： Mitsuo Shimotani , Masahiro Hibino , Kenji Shima
发明人： Mitsuo Shimotani , Masahiro Hibino , Kenji Shima
IPC分类号： G10L11/02 , G10L11/04 , G10L15/00
CPC分类号： G10L25/87 , G10L15/00 , G10L25/90
摘要： A word speech recognition apparatus recognizes a speech inputted to a microphone (11). A feature extracting portion (20) extracts a feature parameter based on a aural signal waveform outputted from the microphone. The feature extracting circuit comprises a pitch cycle extraction circuit (21) for extracting pitch frequency of the speech signal waveform, a digital filter (23) for extracting, as a feature parameter, spectrum data of the speech signal waveform, and a filter coefficient setting circuit (22) for setting a filter coefficient so that a resonance frequency of the digital filter is an integral multiple of the pitch frequency. The feature parameter extracted from the feature extracting circuit is stored in an input pattern memory (3). A recognition processing portion (50) evaluates similarity between the feature parameter stored in advance in a registration pattern memory (4) and the feature parameter stored in the input pattern memory, so that speech recognition processing is made. Improved signal in noise performance results from very narrow bandwidth filters.
摘要翻译：字语音识别装置识别输入到麦克风（11）的语音。特征提取部（20）基于从麦克风输出的听觉信号波形提取特征参数。特征提取电路包括用于提取语音信号波形的音调频率的音调周期提取电路（21），用于提取语音信号波形的频谱数据作为特征参数的数字滤波器（23）和滤波器系数设置电路（22），用于设置滤波器系数，使得数字滤波器的谐振频率是音调频率的整数倍。从特征提取电路提取的特征参数存储在输入图案存储器（3）中。识别处理部分（50）评估预先存储在注册模式存储器（4）中的特征参数与存储在输入模式存储器中的特征参数之间的相似度，从而进行语音识别处理。改进的噪声性能信号来自非常窄的带宽滤波器。

84. 发明授权

US4731845A Device for loading a pattern recognizer with a reference pattern selected from similar patterns 失效
标题翻译：用于加载具有从相似模式中选择的参考模式的模式识别器的装置
公开(公告)号：US4731845A
公开(公告)日：1988-03-15
申请号：US632492
申请日：1984-07-19
申请人： Tomoko Matsuki , Hideo Tanaka
发明人： Tomoko Matsuki , Hideo Tanaka
IPC分类号： G10L11/00 , G07C9/00 , G10L11/02 , G10L15/06 , G10L5/00
CPC分类号： G10L25/87 , G07C9/0015 , G10L15/063
摘要： One reference pattern is selected from three spoken repetitions of the same utterance, the one selected having the highest calculated similarity to the other two, thereby being the most representative.
摘要翻译：从相同话语的三个口头重复中选择一个参考模式，所选择的参考模式与其他两个计算的相似度最高，因此是最具代表性的。

85. 发明授权

US4718088A Speech recognition training method 失效
标题翻译：语音识别训练方法
公开(公告)号：US4718088A
公开(公告)日：1988-01-05
申请号：US593891
申请日：1984-03-27
申请人： James K. Baker , John W. Klovstad , Chin-Hui Lee , Kalyan Ganesan
发明人： James K. Baker , John W. Klovstad , Chin-Hui Lee , Kalyan Ganesan
IPC分类号： G10L11/02 , G10L15/02 , G10L15/06 , G10L15/08 , G10L15/12 , G10L5/00
CPC分类号： G10L25/87 , G10L15/063 , G10L15/083 , G10L15/02 , G10L15/12 , G10L2015/0638 , G10L25/27
摘要： A speech recognition method and apparatus employ a speech processing circuitry for repetitively deriving from a speech input, at a frame repetition rate, a plurality of acoustic parameters. The acoustic parameters represent the speech input signal for a frame time. A plurality of template matching and cost processing circuitries are connected to a system bus, along with the speech processing circuitry, for determining, or identifying, the speech units in the input speech, by comparing the acoustic parameters with stored template patterns. The apparatus can be expanded by adding more template matching and cost processing circuitry to the bus thereby increasing the speech recognition capacity of the apparatus. Template pattern generation is advantageously aided by using a "joker" word to specify the time boundaries of utterances spoken in isolation, by finding the beginning and ending of an utterance surrounded by silence.
摘要翻译：语音识别方法和装置采用语音处理电路，以帧重复率重复地从语音输入中导出多个声学参数。声学参数表示帧时间的语音输入信号。通过将声学参数与存储的模板图案进行比较，多个模板匹配和成本处理电路连同语音处理电路连接到用于确定或识别输入语音中的语音单元的系统总线。可以通过向总线添加更多的模板匹配和成本处理电路来扩展该装置，从而增加装置的语音识别能力。通过使用“小丑”字通过找到由沉默包围的话语的开始和结束来有助于指定孤立地说出的话语的时间边界。

86. 发明授权

US4696041A Apparatus for detecting an utterance boundary 失效
标题翻译：用于检测话语边界的装置
公开(公告)号：US4696041A
公开(公告)日：1987-09-22
申请号：US575383
申请日：1984-01-30
申请人： Tomio Sakata
发明人： Tomio Sakata
IPC分类号： G10L11/00 , G10L11/02 , G10L15/04 , G10L5/00
CPC分类号： G10L25/87
摘要： An utterance boundary detecting apparatus of this invention includes an acoustic processor for generating speech parameter time sequence data according to an input speech signal. The speech parameter time sequence data generated from the acoustic processor is delivered to a buffer memory and noise level determining circuit. The noise level determining circuit calculates the average value of speech parameter values of a background noise corresponding to a silent period when a speech signal is input as words uttered. The apparatus includes a threshold value calculating circuit for calculating an utterance boundary detection threshold value on the basis of an average value calculated by the noise level determining circuit. An utterance boundary detecting circuit generates utterance boundary data on the basis of the threshold value from the threshold value calculating circuit and speech parameter time sequence data in a buffer memory.
摘要翻译：本发明的发声边界检测装置包括声处理器，用于根据输入的语音信号产生语音参数时间序列数据。从声学处理器生成的语音参数时间序列数据被传送到缓冲存储器和噪声电平确定电路。噪声电平确定电路计算当语音信号被输入时与静音时段对应的背景噪声的语音参数值的平均值。该装置包括阈值计算电路，用于根据由噪声电平确定电路计算的平均值来计算发声边界检测阈值。话音边界检测电路根据来自阈值计算电路的阈值和缓冲存储器中的语音参数时间序列数据产生话音边界数据。

87. 发明授权

US4641342A Voice input system 失效
标题翻译：语音输入系统
公开(公告)号：US4641342A
公开(公告)日：1987-02-03
申请号：US590660
申请日：1984-03-19
申请人： Takao Watanabe , Masao Watari
发明人： Takao Watanabe , Masao Watari
IPC分类号： G10L11/02 , G10L15/28 , G10L5/00
CPC分类号： G10L25/87 , G10L15/32
摘要： An input system for a voice recognizer circuit wherein a cue signal is issued to the user to indicate system readiness. A voice detector detects the presence of a voice signal. Control circuitry detects if a voice signal is detected prior to the end of the initial cue signal and, if so, causes a second cue signal to be issued, thereby preventing a loss of voice input. The voice detector can also be used to selectively switch an active one of a plurality of user channels to one of a smaller number of voice recognizers.
摘要翻译：一种用于语音识别器电路的输入系统，其中向用户发出提示信号以指示系统准备就绪。语音检测器检测到语音信号的存在。控制电路检测在初始提示信号结束之前是否检测到语音信号，如果是，则发出第二提示信号，从而防止语音输入的丢失。语音检测器还可以用于选择性地将多个用户信道中的活动的一个切换到较少数量的语音识别器中的一个。

88. 再颁专利

USRE31188E Multiple template speech recognition system 失效
标题翻译：多模板语音识别系统
公开(公告)号：USRE31188E
公开(公告)日：1983-03-22
申请号：US336067
申请日：1981-12-31
申请人： Frank C. Pirz , Lawrence R. Rabiner
发明人： Frank C. Pirz , Lawrence R. Rabiner
IPC分类号： G10L11/02 , G10L15/06
CPC分类号： G10L25/87 , G10L15/063
摘要： A speech analyzer for recognizing an unknown utterance as one of a set of reference words is adapted to generate a feature signal set for each utterance of every reference word. At least one template signal is produced for each reference word which template signal is representative of a group of feature signal sets. Responsive to a feature signal set formed from the unknown utterance and each reference word template signal, a signal representative of the similarity between the unknown utterance and the template signal is generated. A plurality of similarity signals for each reference word is selected and a signal corresponding to the average of said selected similarity signals is formed. The average similarity signals are compared to identify the unknown utterance as the most similar reference word. Features of the invention include: template formation by successive clustering involving partitioning feature signal sets into groups of predetermined similarity by centerpoint clustering, and recognition by comparing the average of selected similarity measures of a time-warped unknown feature signal set with the cluster-derived reference templates for each vocabulary word.

89. 发明授权

US4297528A Training circuit for audio signal recognition computer 失效
标题翻译：音频信号识别计算机训练电路
公开(公告)号：US4297528A
公开(公告)日：1981-10-27
申请号：US073781
申请日：1979-09-10
申请人： Richard D. Beno
发明人： Richard D. Beno
IPC分类号： G06K9/66 , G06T1/00 , G10L11/00 , G10L11/02 , G10L15/06 , G10L1/00
CPC分类号： G10L25/87 , G10L15/063
摘要： A signal pattern recognition system includes a reference pattern memory storing plural reference data patterns against which input patterns are compared for recognition. The reference data patterns are formed by merging training patterns together. Each training pattern, to be accepted for merging, must match the previously merged patterns by a threshold amount. The threshold is automatically varied as the number of previously merged training patterns increases. If a predetermined number of successive training patterns is below the threshold, the training process is repeated, from the beginning, for the reference pattern which generates the errors. The system automatically trains each reference pattern with the same number of training patterns to assure uniformity when input data patterns are compared for recognition.
摘要翻译：信号模式识别系统包括存储多个参考数据模式的参考模式存储器，用于识别输入模式。通过将训练模式合并在一起形成参考数据模式。要被合并的每个训练模式必须与先前合并的模式相匹配，达到阈值。随着先前合并训练模式的数量增加，阈值自动变化。如果预定数量的连续训练模式低于阈值，则从一开始就针对产生错误的参考模式重复训练过程。系统使用相同数量的训练模式自动训练每个参考模式，以确保在输入数据模式进行比较以进行识别时的均匀性。

90. 发明授权

US4292470A Audio signal recognition computer 失效
标题翻译：音频信号识别电脑
公开(公告)号：US4292470A
公开(公告)日：1981-09-29
申请号：US73792
申请日：1979-09-10
申请人： Byung H. An
发明人： Byung H. An
IPC分类号： G10L11/00 , G10L11/02 , G10L15/02 , G10L15/28 , G10L1/00
CPC分类号： G10L25/87 , G10L15/285
摘要： A signal encoder and classifier particularly adapted to speech recognition includes a circularly addressed buffer which is independently addressed by a new data writing address system and a buffered data reading system so that writing and reading of data may be accomplished on a time shared basis. This time shared operation permits serial writing and reading of the pattern data without interrupting income signal storage. The writing data address system addresses the data into the buffer in a circular fashion while the reading data address system utilizes stored addresses identifying the beginning and end of the signal patterns for addressing sequential patterns from the buffer.
摘要翻译：特别适用于语音识别的信号编码器和分类器包括循环寻址的缓冲器，其由新的数据写入地址系统和缓冲的数据读取系统独立地寻址，使得数据的写入和读取可以在时间上共享的基础上完成。这种共享操作允许串行写入和读取模式数据而不中断收入信号存储。写入数据地址系统以循环方式将数据寻址到缓冲器中，而读取数据地址系统利用标识信号模式的开始和结束的存储的地址来从缓冲器寻址顺序模式。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式