公开(公告)号：US5960063A

公开(公告)日：1999-09-28

申请号：US892606

申请日：1997-07-14

申请人： Shingo Kuroiwa ,  Masaki Naito ,  Seiichi Yamamoto ,  Shin-ichi Sakayori ,  Masanobu Fujioka

发明人： Shingo Kuroiwa ,  Masaki Naito ,  Seiichi Yamamoto ,  Shin-ichi Sakayori ,  Masanobu Fujioka

IPC分类号： G10L15/06 ,  G10L15/00 ,  G10L15/20 ,  G10L15/28 ,  G10L21/02 ,  H04M1/27 ,  H04M3/42 ,  H04M3/50

CPC分类号： G10L15/30 ,  G10L15/20 ,  H04M1/271 ,  H04M2201/40 ,  Y10S379/907

摘要： A telephone speech recognition system is provided which has a high level of speech recognition without affecting from various conditions of a telephone line. The system comprises speech analyzers (4) and (5), and reference speech model storages (7) to (9) corresponding to line connection data. A telephone line interface (1) having a line connection data acquisition function analyzes a call received from the telephone line for identifying the country, route, and other information of the call and transmits those line connection data to a line connection data processor (2). The line connection data processor (2) selects one of the acoustic analyzers (4) and (5) in response to the line connection data from the interface (1) and also one of the speech model storages (7) to (9). A speech pattern matcher (11) compares an acoustic vector train output of the selected acoustic analyzer with the speech models given from the selected reference speech model storage for speech recognition.

摘要翻译： 提供了一种电话语音识别系统，其具有高水平的语音识别，而不影响电话线路的各种状况。 该系统包括对应于线路连接数据的语音分析器（4）和（5）以及参考语音模型存储器（7）至（9）。 具有线路连接数据采集功能的电话线接口（1）分析从电话线路接收的用于识别呼叫的国家，路线和其他信息的呼叫，并将这些线路连接数据发送到线路连接数据处理器（2） 。 线路连接数据处理器（2）响应于来自接口（1）的线路连接数据以及语音模型存储器（7）至（9）中的一个，选择声学分析器（4）和（5）中的一个。 语音模式匹配器（11）将所选择的声学分析器的声矢量列输出与从所选择的参考语音模型存储器中给出的用于语音识别的语音模型进行比较。

公开(公告)号：US5740318A

公开(公告)日：1998-04-14

申请号：US534805

申请日：1995-09-27

申请人： Masaki Naito ,  Shingo Kuroiwa ,  Kazuya Takeda ,  Seiichi Yamamoto

发明人： Masaki Naito ,  Shingo Kuroiwa ,  Kazuya Takeda ,  Seiichi Yamamoto

IPC分类号： G10L11/02 ,  G10L15/04 ,  G10L15/14 ,  G10L15/18 ,  G10L15/28 ,  G10L5/06

CPC分类号： G10L25/87 ,  G10L15/193

摘要： A speech endpoint detection unit determines that a first condition is met when a matching score of a partial sentence accepted by a grammatical rule unit is the highest of all partial sentences, and that a second condition is met when a duration time of input speech determined to coincide with a silent standard pattern is longer than a predetermined time. A speech endpoint is determined when the first and second conditions are both met and a speech endpoint detection signal is sent to a word prediction unit and a recognition result output unit. By requiring that both first and second conditions be met, a speech endpoint can correctly be detected even when a long silent period is present in the course of a sentence.

摘要翻译： 当语法规则单元接受的部分句子的匹配分数是所有部分句子中最高的语音终端检测单元确定满足第一条件，并且当输入语音的持续时间确定为 与无声标准模式一致比预定时间长。 当满足第一和第二条件并且语音端点检测信号被发送到字预测单元和识别结果输出单元时，确定语音端点。 通过要求满足第一和第二条件，即使在句子的过程中存在长静音期也能够正确地检测到语音端点。

公开(公告)号：US5425127A

公开(公告)日：1995-06-13

申请号：US896247

申请日：1992-06-10

申请人： Fumihiro Yato ,  Kazuya Takeda ,  Shingo Kuroiwa

发明人： Fumihiro Yato ,  Kazuya Takeda ,  Shingo Kuroiwa

IPC分类号： G10L15/02 ,  G10L15/06 ,  G10L15/10 ,  G10L5/06

CPC分类号： G10L15/10 ,  G10L25/18

摘要： This invention relates to a method of recognizing input speech of many unspecific people. Feature parameters representing both a short-time average spectrum envelope characteristic of the input speech, and regression coefficients obtained from the outputs from a wide-band filter bank. The regression coefficients represent the rough directionality of the characteristic of change in the spectrum of the speech signal. Distance is measured between the feature parameters and standard patterns stored in a storage means. The distance between the feature parameters and the stored pattern which is smallest of all the patterns is found to recognize said input speech.

摘要翻译： 本发明涉及识别许多非特定人的输入语音的方法。 特征参数表示输入语音的短时平均频谱包络特征，以及从宽带滤波器组的输出获得的回归系数。 回归系数表示语音信号频谱变化特性的粗略方向性。 在存储装置中存储的特征参数和标准模式之间测量距离。 发现特征参数与所有图案中最小的存储图案之间的距离识别所述输入语音。

公开(公告)号：US6111937A

公开(公告)日：2000-08-29

申请号：US891997

申请日：1997-07-14

申请人： Shingo Kuroiwa ,  Shin-ichi Sakayori ,  Seiichi Yamamoto ,  Masanobu Fujioka

发明人： Shingo Kuroiwa ,  Shin-ichi Sakayori ,  Seiichi Yamamoto ,  Masanobu Fujioka

IPC分类号： H04M3/42 ,  H04M3/38 ,  H04M3/493 ,  H04M3/51 ,  H04M3/60 ,  H04M1/64 ,  H04M3/00 ,  H04M15/00

CPC分类号： H04M3/382 ,  H04M3/493 ,  H04M3/51 ,  H04M2201/40

摘要： This invention is characterized by providing an apparatus for automatically discriminating service users which can prevent troubles from occurring between the service provider and users because of difference in language. When the service provider has a call from a user, the service provider makes an announcement to the user to request the voicing of a specific keyword. If an answer by the keyword is made from the user, it is determined whether or not it is an answer by the correct keyword. If correct, the call is connected to the service provider. If not correct, how to use the service is announced in another language. Then, the line is disconnected. As a result, a user who is weak in the language used by the service provider can be prevented from being connected to the service provider.

摘要翻译： 本发明的特征在于提供一种用于自动识别服务用户的装置，其可以防止由于语言差异而在服务提供商和用户之间发生麻烦。 当服务提供商具有来自用户的呼叫时，服务提供商向用户通知请求发出特定关键字。 如果从用户进行关键字的回答，则确定是否是正确关键字的答案。 如果正确，则呼叫连接到服务提供商。 如果不正确，如何使用该服务以另一种语言宣布。 然后，线路断开连接。 结果，可以防止由服务提供商使用的语言薄弱的用户连接到服务提供商。