Abstract: An information processing apparatus includes a target storage means, a control means, a decision means, a prediction means and a generating means, in which the decision means, when determining that the system state makes a transition to a state represented by the target value in accordance with the output of the time series of motor signals generated by the generating means, updates a parameter representing the relationship between input and output, which is used by the control means for controlling the system based on the time series of motor signals generated by the generating means and a time series of sensor signals observed in accordance with the output of the time series of motor signals.

Abstract: An information processing apparatus includes a first learning unit adapted to learn a first SOM (self-organization map), based on a first parameter extracted from an observed value, a winner node determination unit adapted to determine a winner node on the first SOM, a searching unit adapted to search for a generation node on a second SOM having highest connection strength with the winner node, a parameter generation unit adapted to generate a second parameter from the generation node, a modification unit adapted to modify the second parameter generated from the generation node, a first connection weight modification unit adapted to modify the connection weight when end condition is satisfied, a second connection weight modification unit adapted to modify the connection weight depending on evaluation made by a user, and a second learning unit adapted to learn the second SOM based on the second parameter obtained when the end condition is satisfied.

Abstract: A sentence generating method and apparatus for generating a joke sentence set having a dual meaning. A predetermined word in a selected first idiom is stored in a W list. A word of analogous sound having the pronunciation same as or analogous with the word in the W list and also being of the same word class as the word in the W list is stored in a RW list. It is then checked whether or not the subject word stored in the P list has a semantic relationship with the word of the analogous sound in the RW list. If the result of check is affirmative, a first joke sentence having the subject word in the P list and composed of the subject word of the P list and the first idiom is generated. A second idiom, corresponding to the first idiom, in which the word in the W list is replaced by the word of the analogous sound in the RW list, is generated, and a second joke sentence composed of the subject word of the P list and the second idiom is generated.

Abstract: A robot apparatus that is able to determine whether to ask a user for help, according to robot's situations. The robot apparatus determines based on an external or internal situation of the robot and its serious level whether to ask for help. When the robot apparatus decides to ask for help, it selects an action asking for help, according to the external or internal situation and the serious level, and takes the action.

Abstract: The present invention relates to a program that allows a conversation to be performed with a user in a natural and human-like manner. Each of response candidate generating modules 54a through 54c generates response candidate data based on its own rule, in response to voice data input from the outside through an input module 51 and a response selecting module 52, and calculates a score of the response candidate data based on the response candidate data, the voice data and internal status data recorded in an internal status recording module 55, and then supplies the score to a response selecting module 52. In response to the supplied score, the response selecting module 52 selects, as response data responsive to the input voice data, one piece of response candidate data, and outputs the one piece of response candidate data through a response output module 53. The present invention is applicable to a voice interactive apparatus mounted on a robot, etc.

Abstract: A learning apparatus for learning time series data in a link model including a plurality of input time series pattern storage networks and a plurality of output time series pattern storage networks with nodes of the input time series pattern storage networks linked to nodes of the output time series pattern storage networks, includes a learning unit for updating in a self-organizing manner each of the plurality of input time series pattern storage networks and updating in a self-organizing manner each of the plurality of output time series pattern storage networks and a link relationship updating unit for updating a link relationship between each node of the output time series pattern storage network and an input winner node, and updating a link relationship between each node of the input time series pattern storage network and an output winner node.

Abstract: A data processing device for processing time-sequence data includes a data extracting unit operable to extract time-sequence data for a predetermined time unit from time-sequence data; and a processing unit operable to obtain scores for nodes of an SOM configured from multiple nodes provided with a spatial array configuration, the scores showing applicability to time-sequence data for a predetermined time unit thereof, wherein the node with the best score thereof is determined to be the winning node which is the node most applicable; wherein the processing unit obtains scores as to the time-sequence data for one predetermined time unit, regarding a distance-restricted node wherein distance from the winning node as to the time-sequence for a predetermined time unit immediately preceding the time-sequence data of one predetermined time unit is within a predetermined distance; and wherein the distance-restricted node with the best the score is determined to be the winning node thereof.

Abstract: A learning apparatus for learning time series data, includes a learning unit for updating, in a self-organizing manner based on an observed value of the time series data, a time series pattern storage network including a plurality of nodes, each node having a time series pattern model representing a time series pattern of the time series data.

Abstract: In a learning system, a learning method, and a robot apparatus, the name of an object is obtained from the user through a dialog with the user. The object is identified based on the detection results of the predetermined plural different features of the object and the learning results of the respective features of the known object previously stored. If it is determined that the above object is a new object based on thus obtained name of the above object, the recognition results respectively based on the respective features of the above object and relation information in that the name of the known object previously stored is connected with the recognition results of the respective features of the known object respectively, the learning of the needed feature of the above object is performed, and relation information on the above object is newly stored.

Abstract: A learning apparatus for learning time series data in a link model including a plurality of input time series pattern storage networks and a plurality of output time series pattern storage networks with nodes of the input time series pattern storage networks linked to nodes of the output time series pattern storage networks, includes a learning unit for updating in a self-organizing manner each of the plurality of input time series pattern storage networks and updating in a self-organizing manner each of the plurality of output time series pattern storage networks and a link relationship updating unit for updating a link relationship between each node of the output time series pattern storage network and an input winner node, and updating a link relationship between each node of the input time series pattern storage network and an output winner node.

Abstract: An information processing apparatus includes a first learning unit adapted to learn a first SOM (self-organization map), based on a first parameter extracted from an observed value, a winner node determination unit adapted to determine a winner node on the first SOM, a searching unit adapted to search for a generation node on a second SOM having highest connection strength with the winner node, a parameter generation unit adapted to generate a second parameter from the generation node, a modification unit adapted to modify the second parameter generated from the generation node, a first connection weight modification unit adapted to modify the connection weight when end condition is satisfied, a second connection weight modification unit adapted to modify the connection weight depending on evaluation made by a user, and a second learning unit adapted to learn the second SOM based on the second parameter obtained when the end condition is satisfied.

Abstract: A robot (1) is provided which includes a situated behaviors layer (SBL) (58). This SBL (58) is formed in the form of a tree structure in which a plurality of schemata (behavior modules) is connected hierarchically in such a matter that the schemata are highly independent of each other for each of them to behave uniquely. A patent schema can define a pattern in which child schemata are connected, such as an OR type pattern in which the child schemata are caused to behave uniquely, AND type pattern in which the plurality of child schemata are caused to behave simultaneously or a SEQUENCE type pattern indicating a sequence in which the plurality of child schemata should behave, thereby permitting to select a behavior pattern of the robot (1). Also, a new child schema can additionally be included in the SBL (58) without having to rewrite the schemata connection in the tree structure, whereby a new behavior or function can be added to the robot (1).

Abstract: A learning apparatus for learning time series data, includes a learning unit for updating, in a self-organizing manner based on an observed value of the time series data, a time series pattern storage network including a plurality of nodes, each node having a time series pattern model representing a time series pattern of the time series data.

Abstract: In a conventional voice dialogue system, there is a case where it is difficult to perform a natural dialogue with the user. Therefore, we designed to perform speech recognition on the user's utterance, to control a dialogue with the user according to a scenario previously given, based on the speech recognition result to generate an answering sentence corresponding to the contents of the user's utterance as the occasion demands, and to perform voice synthesis processing to one sentence in the reproduced scenario or the generated answering sentence.

Abstract: No robot that can make a dialog customized for the user is known to date. According to the invention, various pieces of information appendant to an object are stored as values of the corresponding items of the object and a topic that is related to the topic used in the immediately preceding conversation is selected. Then, an acquisition conversation for acquiring the value of the item of the selected topic or a utilization conversation for utilizing the value of the item of the topic that is already stored is generated as the next conversation. The value acquired by the acquisition conversation is stored as the value of the corresponding item.

Abstract: A translating apparatus includes a speech recognition unit, a machine translation unit, a speech synthesizing unit, and a dialog management unit. The speech recognition unit recognizes input speech, and supplies the result of the speech recognition to the machine translation unit. The machine translation unit translates the result of the speech recognition. When a Japanese-speaking user and an English-speaking user speak to each other, the dialog management unit stores a history of dialogs, and determines based on the dialog history and prestored knowledge whether the result of the translation by the machine translation unit is unnatural compared with the flow of the dialog. When the result of the translation is unnatural, the dialog management unit outputs, to the machine translation unit, information representing the unnatural result. This controls the machine translation unit to perform translation again.

Abstract: This invention realizes a speech communication system and method, and a robot apparatus capable of significantly improving entertainment property. A speech communication system with a function to make conversation with a conversation partner is provided with a speech recognition means for recognizing speech of the conversation partner, a conversation control means for controlling conversation with the conversation partner based on the recognition result of the speech recognition means, an image recognition means for recognizing the face of the conversation partner, and a tracking control means for tracing the existence of the conversation partner based on one or both of the recognition result of the image recognition means and the recognition result of the speech recognition means. The conversation control means controls conversation so as to continue depending on tracking of the tracking control means.

Abstract: Icons indicating operating conditions of each of nodes (computers) constituting a system are individually displayed in the form of a loop. When the total number of the nodes exceeds a total number B of individual displays, nodes corresponding to B are displayed individually, and the rest are displayed in the form of a node collective icon. When a forward button is clicked, a display position of each of the node icons revolves clockwise one by one, a node that has been in the node collective icon is displayed individually at a position of a node icon, and an node icon at a position of a node icon is contained in the node collective icon. By clicking buttons a clockwise continuous revolving display, a counterclockwise revolving display and a counterclockwise continuous revolving display are performed. Node icons are highlighted so as to indicate abnormal conditions of the corresponding nodes.

Abstract: Conventional robot apparatus etc. can not perform name-learning naturally. Learning the name of an object is performed such a manner that the name of a target object is obtained through dialog with a human being, the name is stored in association with plural items of different characteristic data detected for the target object, and a new object is recognized based on the stored data and associative information, the name and characteristic data of the new person are obtained and this associative information is stored.

Abstract: A robot apparatus that is able to determine whether to ask a user for help, according to robot's situations. The robot apparatus determines based on an external or internal situation of the robot and its serious level whether to ask for help. When the robot apparatus decides to ask for help, it selects an action asking for help, according to the external or internal situation and the serious level, and takes the action.