Sony Corporation

Sony Computer Entertainment Inc.

Tsinghua University

Sony France S.a.

Makoto Akabane

Kenichiro Kobayashi

Nobuhide Yamazaki

Tomoaki Nitta

Masao Watari

Erika Kobayashi

Keiichi Yamada

Kazuo Ishii

Hiroaki Ogawa

Yasuharu Asano

Yasuhiko Kato

Tetsuya Kagami

Hiroshi Kakuda

Masato Shimakawa

Akira Tange

Miyuki Tanaka

Hajime Yano

Junichi Kudo

Goro Shiraishi

Kazuo Watanabe

Katsuki Minamino

Yoshikazu Takahashi

Masanori Omote

Yusuke Iwahashi

Yoichiro Sako

Junichi Rekimoto

Masahiro Fujita

Ruxin Chen

Hiroaki Kitano

Toru Yamamoto

Masashi Takeda

Satoshi Fujimura

Akira Sassa

Toshitada Doi

Atsuo Hiroe

Fukuharu Sudo

Masashi Ohkubo

Koji Kageyama

Katsu Saito

Hitoshi Honda

Toshiyuki Kumakura

Eiji Yamamoto

Hisao Nishioka

Rika Horinaka

Toshihiko Waku

Atsunobu Hiraiwa

Fang Zheng

Pierre Yves Oudeyer

Xi Xiao

Hideki Kishi

Satoko Ikezawa

Bao Hongchang

Linquan Liu

Kazunori Sasano

Zhan You

Makoto Inoe

Wenxiao Cao

The present invention relates to a method for modeling a common-language speech recognition, by a computer, under the influence of multiple dialects and concerns a technical field of speech recognition by a computer. In this method, a triphone standard common-language model is first generated based on training data of standard common language, and first and second monophone dialectal-accented common-language models are based on development data of dialectal-accented common languages of first kind and second kind, respectively. Then a temporary merged model is obtained in a manner that the first dialectal-accented common-language model is merged into the standard common-language model according to a first confusion matrix obtained by recognizing the development data of first dialectal-accented common language using the standard common-language model. Finally, a recognition model is obtained in a manner that the second dialectal-accented common-language model is merged into the temporary merged model according to a second confusion matrix generated by recognizing the development data of second dialectal-accented common language by the temporary merged model. This method effectively enhances the operating efficiency and admittedly raises the recognition rate for the dialectal-accented common language. The recognition rate for the standard common language is also raised.

Method and system for modeling a common-language speech recognition, by a computer, under the influence of a plurality of dialects

In a music information processing apparatus, a characteristic quantity acquiring unit reads out a music file from a music tune file storage unit and acquires a characteristic quantity of a music tune. A class value providing unit provides a class value for each music tune based on the characteristic quantity. The distribution acquiring unit forms a music tune assembly based on bibliography information read from the music tune file by a bibliography information acquiring unit, and acquires the distribution information of class values for each music tune assembly. The distribution information is stored into an analysis result storage unit, and an evaluation unit reads out distribution information based on user input, and evaluates the similarity between music tune assemblies or between a tune and a music tune assembly. An output unit recommends or reproduces a music tune assembly or a music tune according to the preference of a user.

Music information processing apparatus, music delivering system, and music information processing method that can satisfy a request from a user

The text format of input data is checked, and is converted into a system-manipulated format. It is further determined if the input data is in an HTML or e-mail format using tags, heading information, and the like. The converted data is divided into blocks in a simple manner such that elements in the blocks can be checked based on repetition of predetermined character patterns. Each block section is tagged with a tag indicating a block. The data divided into blocks is parsed based on tags, character patterns, etc., and is structured. A table in text is also parsed, and is segmented into cells. Finally, tree-structured data having a hierarchical structure is generated based on the sentence-structured data. A sentence-extraction template paired with the tree-structured data is used to extract sentences.

Document processing apparatus, document processing method, document processing program and recording medium

The present invention is intended to provide a text-to-speech synthesis apparatus, including a storage for storing phoneme data of a plurality of speakers; a selector for selecting one of the plurality of speakers in accordance with an operation performed by a user; a searcher for searching the storage for phoneme data of the speaker selected by the selector; a text-to-speech synthesis processor for linking the phoneme data of the speaker retrieved by the searcher to convert input data into a synthetic speech; and a fee-charge controller for controlling a fee-charge operation for the user in accordance with the phoneme data selected by the selector. Consequently, the user can perform text-to-speech synthesis on the desired input data such as drama data by use of the obtained phoneme data.

Text-to-speech synthesis system

The emotion is to be added to the synthesized speech as the prosodic feature of the language is maintained. In a speech synthesis device, a language processorgenerates a string of pronunciation marks from the text, and a prosodic data generating unitcreates prosodic data, expressing the time duration, pitch, sound volume or the like parameters of phonemes, based on the string of pronunciation marks. A constraint information generating unitis fed with the prosodic data and with the string of pronunciation marks to generate the constraint information which limits the changes in the parameters to add the so generated constraint information to the prosodic data. A emotion filter, fed with the prosodic data, to which has been added the constraint information, changes the parameters of the prosodic data, within the constraint, responsive to the feeling state information, imparted to it. A waveform generating unitsynthesizes the speech waveform based on the prosodic data the parameters of which have been changed.

Method and apparatus for speech synthesis, program, recording medium, method and apparatus for generating constraint information and robot apparatus

Various sensors detect conditions outside a robot and an operation applied to the robot, and output the results of detection to a robot-motion-system control section. The robot-motion-system control section determines a behavior state according to a behavior model. A robot-thinking-system control section determines an emotion state according to an emotion model. A speech-synthesizing-control-information selection section determines a field on a speech-synthesizing-control-information table according to the behavior state and the emotion state. A language processing section analyzes in grammar a text for speech synthesizing sent from the robot-thinking-system control section, converts a predetermined portion according to a speech-synthesizing control information, and outputs to a rule-based speech synthesizing section. The rule-based speech synthesizing section synthesizes a speech signal corresponding to the text for speech synthesizing.

Speech synthesizing apparatus, speech synthesizing method, and recording medium using a plurality of substitute dictionaries corresponding to pre-programmed personality information

A communication apparatus includes a storage unit, a receiving unit and a searching unit, all acting on speaker voice data. A communication system includes an accumulating unit, a requesting unit, a transmitting unit and a receiving unit, all acting on synthetic speech data.

Communication apparatus and system acting on speaker voices

Document processing apparatus, document processing method, document processing program, and recording medium

The present invention relates to a voice output apparatus capable of, in response to a particular stimulus, stopping outputting a voice and outputting a reaction. The voice output apparatus is capable of outputting a voice in a natural manner. A rule-based synthesizerproduces a synthesized voice and outputs it. For example, when a synthesized voice 'Where is an exit' was produced and outputting of the synthesized voice data has proceeded until 'Where is an e' has been output, if a user taps a robot, then a reaction generatordetermines, by referring to a reaction database, that a reaction voice &#x201c;Ouch!&#x201d; should be output in response to being tapped. The reaction generatorthen controls an output controllerso as to stop outputting the synthesized voice &#x201c;Where is an exit?&#x201d; and output the reaction voice &#x201c;Ouch!&#x201d;. Thereafter, the reaction generatorcontrols the read pointer of a buffercontrolled by the read controllersuch that the outputting of the synthesized voice is resumed from the point at which the outputting was stopped. Thus, the synthesized voice &#x201c;Where is an e, Ouch!, xit?&#x201d; is output.

Speech output apparatus

A clientincludes a transmission unitfor transmitting the input speech information over a network to a server systemand an output unitfor receiving the contents selection information from the server systemover the network to output the received information. The server systemincludes a prepared information storage unitfor memorizing one or more pieces of the preparation information pertinent to each contents, from one contents to another, and an information preparing serverfor preparing the contents selection information based on the speech information received from the clientover the network and on the preparation information to send the so-prepared contents selection information to the clientover the network.

Content selection system

In a synchronization control apparatus, a voice-language-information generating section generates the voice language information of a word which a robot utters. A voice synthesizing section calculates phoneme information and a phoneme continuation duration according to the voice language information, and also generates synthesized-voice data according to an adjusted phoneme continuation duration. An articulation-operation generating section calculates an articulation-operation period according to the phoneme information. A voice-operation adjusting section adjusts the phoneme continuation duration and the articulation-operation period. An articulation-operation executing section operates an organ of articulation according to the adjusted articulation-operation period.

Synchronization control apparatus and method, and recording medium

A sentence or a singing is to be synthesized with a natural speech close to the human voice. To this end, singing metrical data are formed in a tag processing unitin a singing synthesis unitin a speech synthesis apparatusbased on singing data and an analyzed text portion. A language analysis unitperforms language processing on text portions other than the singing data. As for a text portion registered in a natural metrical dictionary, as determined by this language processing, corresponding natural metrical data is selected and its parameters are adjusted in a metrical data adjustment unitbased on phonemic segment data of a phonemic segment storage unitin the metrical data adjustment unit. As for a text portion not registered in the natural metrical dictionary, a phonemic symbol string is generated in a natural metrical dictionary storage unit, after which metrical data are generated in a metrical generating unit. A waveform generating unitconcatenates necessary phonemic segment data, based on the natural metrical data, metrical data and the singing metrical data to generate speech waveform data.

Speech synthesis method and apparatus, program, recording medium and robot apparatus

Growing community of inventors

Tokyo, Japan

Makoto Akabane