专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

JPH11122114A CODE BOOK GENERATOR, CODE BOOK GENERATING METHOD, VECTOR QUANTIZATION DEVICE AND VECTOR QUANTIZATION METHOD 失效
公开(公告)号：JPH11122114A
公开(公告)日：1999-04-30
申请号：JP28330997
申请日：1997-10-16
申请人： SONY CORP
发明人： ECHIZEN TAKAMASA , OGAWA HIROAKI , WATARI MASAO
IPC分类号： G10L15/14 , G06K9/62 , G10L19/00 , H03M7/30 , G10L3/00 , G10L9/18
摘要： PROBLEM TO BE SOLVED: To generate a code book used to reduce a vector quantization error by a memory with a small memory capacity in a short time. SOLUTION: Each distance between respective nodes arranged on a Kohonen self-organization characteristic map and characteristic vectors A(A1 -An N) as learning samples is calculated. A survival node that is a node having the shortest distance from the learning samples is detected. Then vectors assigned to nodes included in the vicinity of a phase as a range of a prescribed distance from the survival node are updated and the self-organization characteristic map obtained from them is used for a code book where the vectors assigned to the nodes are used for code vectors.

2. 发明专利

JPH0934895A TRANSLATION SYSTEM 失效
公开(公告)号：JPH0934895A
公开(公告)日：1997-02-07
申请号：JP17862395
申请日：1995-07-14
申请人： SONY CORP
发明人： ISHII KAZUO , ASANO KOJI , OGAWA HIROAKI , KATO YASUHIKO , KAGAMI TETSUYA , WATARI MASAO , AKAHA MAKOTO , TANAKA MIYUKI , TSUNODA HIROSHI
IPC分类号： G06F17/28 , H04L12/00
摘要： PROBLEM TO BE SOLVED: To simply configure the translation system among multi-languages at a low cost by translating a prescribed language inputted in each translation device into a common language and sending the translation result into the common language to other translation device. SOLUTION: Upon receipt of a voice in a language A, a voice recognition section 201 recognizes the voice and provides an output of the voice recognition result to a translation section 202. Then the translation section 202 translates the voice recognition result for the voice in the language A into a language C used in common for all voice translation devices 200, 210, and provides the result of translation to a MODEM 205. Furthermore, the MODEM 205 sends the translation result into the common language C to other voice translation device 210 other than the voice translation device 200 via a radio channel or a wired line. Moreover, the MODEM 205 receives the translation result into the common language C and gives it to the translation section 204. Then the translation section 204 translates the translation result into the common language from the MODEM 205 into the language A and provides an output of the translation result to a voice synthesis section 203.

3. 发明专利

JPH08123909A CHARACTER RECOGNITION DEVICE 失效
公开(公告)号：JPH08123909A
公开(公告)日：1996-05-17
申请号：JP26566794
申请日：1994-10-28
申请人： SONY CORP
发明人： OMOTE MASANORI , WATARI MASAO , TANAKA MIYUKI , ISHII KAZUO , KATO YASUHIKO , OGAWA HIROAKI , WATANABE KAZUO , MINAMINO KATSUKI , HONDA HITOSHI
IPC分类号： G06K9/62
摘要： PURPOSE: To surety recognize a character in a snort time by extracting features of a picture pattern based on specific mapping to recognize the character corresponding to the picture pattern. CONSTITUTION: An input means (input part) 101 which inputs the picture pattern, a recognition means (feature extraction part) 102 which extracts features of the picture pattern based on mapping determined by the mapping determination method, where a function fm (X) of the m-th component of mapping F is expressed by the linear sum of products between Lm functions g1m (X) and coefficients C1m , to recognize the character corresponding to the picture pattern, and an output means (output part) 103 which outputs the recognition result are provided. Features of the inputted picture pattern are extracted based on mapping determined by the mapping determination method to recognize that this picture pattern corresponds to a prescribed character. Consequently, the corresponding character is surely recognized from the inputted picture pattern in a short time independently of the initial state of the device at the time of learning.

4. 发明专利

JPH08123908A CHARACTER RECOGNITION DEVICE 失效
公开(公告)号：JPH08123908A
公开(公告)日：1996-05-17
申请号：JP26497894
申请日：1994-10-28
申请人： SONY CORP
发明人： HONDA HITOSHI , WATARI MASAO , TANAKA MIYUKI , ISHII KAZUO , KATO YASUHIKO , OGAWA HIROAKI , OMOTE MASANORI , WATANABE KAZUO , MINAMINO KATSUKI
IPC分类号： G06K9/62
摘要： PURPOSE: To obtain the correct recognition result with an easy processing without learning for a long time neither depending upon initial values used for learning by recognizing a character based on the conversion result of a conversion means which performs conversion by a specific function. CONSTITUTION: An extracting means which extracts a feature quantity X from an inputted character and a conversion means which converts the extracted feature quantity X by a function fm (X) where the m-th component is expressed by the linear sum of products between Lm functions gm K (X) and coefficients Cm K are provided, and the character is recognized based on the conversion result of the conversion means. In this case, the character can be a figure. That is, the feature quantity X is extracted from the inputted character and is converted by the function fm (X) where the m-th component is expressed by the linear sum of products between Lm functions gm K (X) and coefficients Cm K, and the character is recognized based on the conversion result. Consequently, the correct recognition result is obtained by the easy processing without learning for a long time neither depending upon initial values used for learning.

5. 发明专利

JPH08123489A SPEECH PITCH FREQUENCY DETECTING METHOD 失效
公开(公告)号：JPH08123489A
公开(公告)日：1996-05-17
申请号：JP26495094
申请日：1994-10-28
申请人： SONY CORP
发明人： AKAHA MAKOTO , WATARI MASAO , KAGAMI TETSUYA , ASANO KOJI
IPC分类号： G10L25/90 , G10L9/00
摘要： PURPOSE: To accurately and speedily detect the pitch frequency of a speech signal. CONSTITUTION: A speech input part 1 converts the speech signal from analog to digital and a speech segmentation part 2 segments a specific speech section. A GGM conversion part 3 converts the data outputted from the speech segmentation part 2 into data on a pitch period position by using mapping. In this this mapping, a function fm (X) of X of an (m)th component is represented as the linear sum of the products of Lm base functions g1m (X) and a coefficient c1m . A pitch cycle position detection part 4 detects a cycle at a position where the value of the pitch cycle detection parameter outputted by the GGM conversion part 3 is close to 1 as a pitch cycle.

6. 发明专利

JPH07311591A VOICE RECOGNITION DEVICE AND NAVIGATION SYSTEM 失效
公开(公告)号：JPH07311591A
公开(公告)日：1995-11-28
申请号：JP10313094
申请日：1994-05-18
申请人： SONY CORP
发明人： WATARI MASAO , ISHII KAZUO , KATO YASUHIKO , OGAWA HIROAKI , OMOTE MASANORI , WATANABE KAZUO , MINAMINO KATSUKI
IPC分类号： G09B29/00 , G10L15/00 , G10L15/18 , G10L15/28 , G10L3/00
摘要： PURPOSE:To conduct a practical voice recognition using a relatively small-scale voice recognition means by conducting the voice recognition employing the vocabulary selected from a dictionary as the voice recognition object vocabulary. CONSTITUTION:When a voice is inputted into a microphone 1, the voice is made into electrical sound signals and outputted to a voice recognition section 2. In the section 2, a voice recognition is conducted for the sound signals from the microphone 1 employing the vocabulary stored in a recognition object word dictionary memory 6 as the voice recognition object vocabulary. Namely, in the section 2, the voice recognition is conducted employing the vocabulary currently stored in the memory 6 only as the object. Therefore, even though the number of the vocabularies registered in a large-scale dictionary 21 is very large, a practical voice recognition is conducted by changing the vocabularies stored in the memory 6 using the entire vocabularies registered in the dictionary 21 as the virtual objects.

7. 发明专利

JPH07230293A VOICE RECOGNITION DEVICE 失效
公开(公告)号：JPH07230293A
公开(公告)日：1995-08-29
申请号：JP2004494
申请日：1994-02-17
申请人： SONY CORP
发明人： ISHII KAZUO , WATARI MASAO , KATO YASUHIKO , OGAWA HIROAKI , OMOTE MASANORI , WATANABE KAZUO , MINAMINO KATSUKI
IPC分类号： G10L11/02 , G10L15/04 , G10L3/00
摘要： PURPOSE:To improve the degree of freedom of ultering of a speaker and the recognition rate of voice. CONSTITUTION:In a control section 5, the silence interval of the voice detected by a voice interval detection section 5 and prescribed set times T1 and T2 are respectively compared (where T1

8. 发明专利

JPH0730675A VOICE ADAPTER 失效
公开(公告)号：JPH0730675A
公开(公告)日：1995-01-31
申请号：JP16944193
申请日：1993-07-09
申请人： SONY CORP
发明人： KATO YASUHIKO , WATARI MASAO , TANAKA MIYUKI , TAJIMA MIE
IPC分类号： G10L15/00 , H04M1/00 , H04M11/00 , H04Q9/00 , G10L3/00 , H04Q7/38
摘要： PURPOSE:To provide voice remote control by using a telephone terminal. CONSTITUTION:When the hand set of a telephone set 9 (slave unit) is off-hooked and the operating instruction of an equipment desired to be remotely controlled is sounded, the voice signal (operating instruction) is supplied from a telephone set 10 (base unit) to a voice recognizing device 4. At the voice recognizing device 4, the voice of this voice signal is recognized by referring to a standard pattern stored in a built-in standard pattern memory 4A, and the recognized result (operating instruction) is outputted to a controller 6. At the controller 6, a remote control signal corresponding to the operating instruction outputted from the voice recognizing device 4 is read from a built-in memory 6A and supplied to an infrared remote control signal emitter 2. At the infrared remote control signal emitter 2, infrared rays corresponding to the remote control signal supplied from the controller 6 are sent to an equipment defined as a control target.

9. 发明专利

JPH06214593A WORD PROCESSOR 失效
公开(公告)号：JPH06214593A
公开(公告)日：1994-08-05
申请号：JP702593
申请日：1993-01-19
申请人： SONY CORP
发明人： WATARI MASAO , TANAKA MIYUKI , KATO YASUHIKO
IPC分类号： G10L15/10 , G10L15/00 , G10L15/18 , G10L15/28 , G10L3/00
摘要： PURPOSE:To outstandingly simplify the registering operations of a user dictionary and to simplify retrieval of the user dictionary. CONSTITUTION:The word processor has a voice recognition device 6 which recognizes the voice from a microphone 5, a character string specifying function section 8 which specifies a character string in a sentence already inputted and a user dictionary memory 4 which stores registered character strings. The memory 4 registers specified character strings or the character strings which are in a half-decided stage and simultaneously registers voice corresponding to the character strings and a calling for registered character strings is performed by voice after the registering. Moreover, a voice recognition receiving condition is decided by a voice input condition switch 10 and a next candidate is retrieved by a voice recognition next candidate key 9 when an erroneous voice recognition is made.

10. 发明专利

JPH0632008B2 失效
公开(公告)号：JPH0632008B2
公开(公告)日：1994-04-27
申请号：JP1584685
申请日：1985-01-30
申请人： SONY CORP
发明人： WATARI MASAO , SAKO YOICHIRO , AKAHA MAKOTO , HIRAIWA ATSUNOBU
IPC分类号： G10L11/00 , G10L15/10 , G10L3/00

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