专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

51. 发明专利

JPH04125596A SPEECH RECOGNIZING DEVICE 失效
公开(公告)号：JPH04125596A
公开(公告)日：1992-04-27
申请号：JP24407490
申请日：1990-09-17
申请人： HITACHI LTD
发明人： MURAYAMA HIDEKI , AMANO AKIO
IPC分类号： G10L11/00 , G10L15/06 , G10L15/10 , G10L15/18
摘要： PURPOSE:To improve recognition accuracy at high speed by performing recognition processing on a recognition target satisfying a condition with respect to the continuous length of a unit template. CONSTITUTION:The feature quantity analyzed by a voice analysis part 102 of voice converted to an electrical signal is extracted. Meanwhile, the text candidate of the recognition target is generated at a text generating part 106 by using the connection rule of the word of a word storage part 103 and a grammar storage part 104, and the minimum voice emission time and the maximum voice emission time of a continuous length condition storage part 105. At a scoring part 107, the recognition target of the word is decided by input feature quantity and the maximum continuous length, and scoring is applied to the feature quantity of the text candidate, and a recognition result is outputted from a selection part 108. In such a way, the recognition accuracy and speed can be improved by setting each unit template only when it is provided with the most likelihood.

52. 发明专利

JPH0272397A SPEECH RECOGNITION DEVICE 失效
公开(公告)号：JPH0272397A
公开(公告)日：1990-03-12
申请号：JP22231388
申请日：1988-09-07
申请人： HITACHI LTD
发明人： AMANO AKIO , ICHIKAWA HIROSHI , HATAOKA NOBUO
IPC分类号： G06F3/16 , G06F15/18 , G06N3/00 , G10L15/02 , G10L15/08 , G10L15/10 , G10L15/16 , G10L15/28
摘要： PURPOSE:To cope flexibly with the fuzziness that a speech has originally and to obtain high recognizability by employing processing by a neural network or fuzzy processing. CONSTITUTION:An input speech analysis part 1, a feature extraction part 2, and a decision part for the category of an input speech are provided and the feature extraction part 2 consists of the neural network. This neural network is increased in the number of constituent layers to realize optional mapping (input/output relation) in principle and this optional mapping is realized concretely by supplying a sample of output data corresponding to input data and learning according to proper learning algorithm. Consequently, it is decided finally which phoneme the input speech is including both self-learning and combination which is learnt previously and fixed.

53. 发明专利

JPH01276200A SPEECH SYNTHESIZING DEVICE 失效
公开(公告)号：JPH01276200A
公开(公告)日：1989-11-06
申请号：JP10391288
申请日：1988-04-28
申请人： HITACHI LTD
发明人： AMANO AKIO , ICHIKAWA HIROSHI , ASAKAWA YOSHIAKI , HATAOKA NOBUO
IPC分类号： G10L13/00
摘要： PURPOSE:To improve the quality by filtering a synthetic speech through a postprocessing filter consisting of a neutral network. CONSTITUTION:In a learning mode, an original speech serving as learning data is inputted from an input part 1 and held in a buffer 2. The initial value of a coupling coefficient is set to a proper value for a multilayered neutral network(NN) 7. Further, a changeover switch 4 is connected so that the output of an analysis part 3 is inputted directly to a coupling part 6. The original speech is read out of the buffer 2, analyzed and converted by the analysis part 3 into analytic parameters, which are inputted to the synthesis part 6 through the changeover switch 4. The synthesis part 6 synthesizes a speech according to the input parameters and the synthesized speech is inputted to the NN 7, and processed and outputted. The output of the NN 7 is inputted to a learning control part 8, and the original speech data is inputted from the buffer 2 to the NN 7 to control the learning of the NN 7, thereby varying the coupling coefficient.

54. 发明专利

JPH01255925A INTERACTIVE TYPE INFORMATION RETRIEVING DEVICE BY VOICE 失效
公开(公告)号：JPH01255925A
公开(公告)日：1989-10-12
申请号：JP8292888
申请日：1988-04-06
申请人： HITACHI LTD
发明人： HATAOKA NOBUO , AMANO AKIO , ICHIKAWA HIROSHI
IPC分类号： G10L11/00 , G06F17/30 , G10L13/00 , G10L15/00 , G10L15/18 , G10L15/28
摘要： PURPOSE:To improve a recognition performance by fetching only the combination of the related word candidate while using the information being described the relation between words already, in plural recognizing candidates obtained by recognizing an inputted voice. CONSTITUTION:The inputted voice is used as plural recognizing candidates by a voice recognizing part 2, and the word series for retrieving are selected by a word series selecting part 3 out of them. The relation between key words concerning the information retrieved beforehand is described in a concept network table 8. A retrieving processing part 4 prepares the retrieving conditions sentence from a word series for retrieving and a concept network, collates this with a concept relation network, selects the contents to be retrieved is selected and the information is retrieved. Thus, a user can retrieve the information even from the ambiguous memory while the user freely interacts with a system, and the recognition performance can be improved.

55. 发明专利

JPS63148299A WORD VOICE RECOGNITION EQUIPMENT 失效
公开(公告)号：JPS63148299A
公开(公告)日：1988-06-21
申请号：JP29703286
申请日：1986-12-12
申请人： HITACHI LTD
发明人： AMANO AKIO , HATAOKA NOBUO
IPC分类号： G10L11/00 , G10L15/10 , G10L15/18 , G10L15/28

56. 发明专利

JPS6188330A Edition processing system of speech typewriter 失效
标题翻译：语音类型编辑处理系统
公开(公告)号：JPS6188330A
公开(公告)日：1986-05-06
申请号：JP20970884
申请日：1984-10-08
申请人： Hitachi Ltd
发明人： YABUUCHI SHIGERU , AMANO AKIO
IPC分类号： G10L11/00 , G06F3/16 , G10L15/00
摘要： PURPOSE: To simplify edition processing by providing a circuit storing the relation between a single syllable code outputted as the result of recognition of an input voice and a represented character code on a display device to apply movement or correction of a cursor at each input single syllable.
CONSTITUTION: When the user utters a Japanese sentence while sectioning it, a voice recognition section 101 recognizes single syllables based on a voice signal from a microphone 100. The output is read by an edition section 104 every time the output is stored in an input syllable buffer 102. Then the display code on a display device 107 of the said syllable code is obtained from a syllable code representation table 103 and the obtained represented character code is written in an input Kana (Japanese syllabary) character series display pattern buffer 105. Then the content of the buffer 105 is displayed on a CRT 107 via a display control section 106 and the correction/insertion/deletion of the Japanese word sentence displayed via a cursor control section 111 and a cursor register 112 is executed in the unit of single syllables.
COPYRIGHT: (C)1986,JPO&Japio
摘要翻译：目的：通过提供一种电路来简化版本处理，该电路存储作为输入声音的识别结果输出的单个音节代码与显示装置上的表示字符码之间的关系，以在每个输入单音节上应用光标的移动或校正。构成：当用户在分割时发出日语句子时，语音识别部分101基于来自麦克风100的语音信号识别单个音节。每当输出存储在输入音节中时，输出被编辑部分104读取然后，从音节代码表103获得所述音节代码的显示装置107上的显示代码，并将所获得的表示的字符代码写入输入的假名（日语音节）字符串显示模式缓冲器105中。然后缓冲器105的内容经由显示控制部分106显示在CRT107上，经由光标控制部分111和光标寄存器112显示的日语单词的校正/插入/删除以单个音节为单位执行。

57. 发明专利

JP2007088803A Information processor 审中-公开
标题翻译：信息处理器
公开(公告)号：JP2007088803A
公开(公告)日：2007-04-05
申请号：JP2005274885
申请日：2005-09-22
申请人： Hitachi Ltd , 株式会社日立製作所
发明人： TOGAMI MASATO , AMANO AKIO , SHINJO HIROSHI , ISHIBASHI ATSUSHI
IPC分类号： H04N17/00 , G06T1/00 , G10L15/00 , G10L15/24 , G10L15/28 , H04N7/173 , H04N21/442
摘要： PROBLEM TO BE SOLVED: To provide a program presentation device by which a system discriminates such a funny television program that a user laughs by voice and such an interesting and involved television program that a human is surprised by voice, etc. SOLUTION: In a media processor, user detection is performed based on image information acquired by an image acquisition part to estimate a face position of the user. In addition, a voice section is detected by voice data recorded with a microphone array. When the sound source direction of the detected voice section matches to the face position, an object for utterance is considered as the media processor. Then, utterance of the voice section is judged as the utterance uttered because the user gets interested in a video image displayed in the media processor and a video section at the same time as that of the voice section is judged as the video section in which the user gets interested. COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：提供一种程序呈现装置，通过该程序呈现装置，系统区分用户通过语音笑出的有趣的电视节目以及人类惊讶于语音等的有趣且有意义的电视节目。
解决方案：在媒体处理器中，基于由图像获取部分获取的图像信息来执行用户检测以估计用户的脸部位置。此外，通过用麦克风阵列记录的语音数据来检测语音部分。当检测到的声音部分的声源方向与脸部位置匹配时，将话音对象视为媒体处理器。然后，由于用户对媒体处理器中显示的视频图像感兴趣，所以语音部分的发音被判定为发声，并且与语音部分同时的视频部分被判断为视频部分，其中，用户感兴趣版权所有（C）2007，JPO＆INPIT

58. 发明专利

JP2001147698A METHOD OF GENERATING PSEUDO WORD FOR VOICE RECOGNITION AND VOICE RECOGNITION DEVICE 失效
公开(公告)号：JP2001147698A
公开(公告)日：2001-05-29
申请号：JP33093699
申请日：1999-11-22
申请人： HITACHI LTD
发明人： KUJIRAI TOSHIHIRO , OBUCHI YASUNARI , AMANO AKIO
IPC分类号： G10L15/28 , G10L15/08
摘要： PROBLEM TO BE SOLVED: To easily provide an out-of-candidate speaking rejection function for a recognition system which doesn't have this function. SOLUTION: Plural pseudo-words are generated with respect to a keyword, and reception or rejection of speaking is decided in accordance with how the degrees of similarity of pseudo-words for speaking are ranked.

59. 发明专利

JPH11212592A PATTERN RECOGNITION DEVICE AND STANDARD PATTERN GENERATING METHOD 失效
公开(公告)号：JPH11212592A
公开(公告)日：1999-08-06
申请号：JP1230198
申请日：1998-01-26
申请人： HITACHI LTD
发明人： AMANO AKIO
IPC分类号： G10L15/06 , G10L15/14 , G10L3/00
摘要： PROBLEM TO BE SOLVED: To provide a standard pattern which has a high recognition degree by measuring how much temporarily generated probability distributions by classes overlap with each other, and correcting distributions which overlap with each other and reducing their overlap. SOLUTION: As for the completed probability distributions of the respective classes, how much the classes overlap with each other is measured (302). Denoting the number of the classes to be recognized as N, (N2-N)/2 pairs are present, so overlaps of the distributions are measured as to all the pairs. When a measured value exceeds a previously set threshold value, it is considered that the distributions overlap large and the number of distributions exceeding the threshold value is counted and denoted as K (303). Then it is decided whether or not the count value K is larger than the previously set threshold value Kth (304) and when the K is larger than Kth, the process is finished, but when not, those probability distributions are selected as distributions to be corrected (305) and corrected (306).

60. 发明专利

JPH1195789A VOICE RECOGNITION SYSTEM AND SPEAKER ADAPTIVE METHOD IN THE SAME 失效
公开(公告)号：JPH1195789A
公开(公告)日：1999-04-09
申请号：JP25984497
申请日：1997-09-25
申请人： HITACHI LTD
发明人： HATAOKA NOBUO , OBUCHI YASUNARI , ODAKA TOSHIYUKI , AMANO AKIO , EJIRI MASAKAZU , OBA SHINYA
IPC分类号： G10L15/10 , G10L15/00 , G10L15/24 , G10L15/28 , G10L3/00
摘要： PROBLEM TO BE SOLVED: To enable anyone to use a voice recognition system effectively by performing a voice recognition while using the acoustic model of the voice recognition system suited for the feature of a speaker estimated in accordance with a detected physical feature information of the speaker to compensating the degradation of a recognition rate to be caused by depending on feature of the voice of a user. SOLUTION: The feature of the user is extracted from a voice 100 in the personal information extracting part 310 of the voice of the user by using the voice 100 or measured results of a mat type scale 160 and/or a sheet position detecting meter 170 as input information. Moreover, when the measured information of the mat type scale 160 and/or the sheet potion detecting meter are used as the input information, the information are converted into information of a weight or a stature in a physical feature converting part 320 and these are converted into the feature of the voice of the user in a voice feature conversion part 330. Then, a speaker adaptation is executed in a speaker adaptation part 350 for acoustic model so that an acoustic model to be used in the voice recognition becomes the acoustic model suited for the user based on the personal information detected in this manner.

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