JP2005258597A - Interaction apparatus and language data conversion method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve user confidentiality protection by concealing the identity of a sending user assumable from language data and video data received by a receiving interaction apparatus. <P>SOLUTION: The interaction apparatus has an attributive phrase data holding part holding an attributive phrase data group corresponding to common language information for each kind of attribute information, and common language-attribute correspondence tables associating a corresponding attribute information group with each piece of common language information; and comprises an input/output control part for receiving language data and attribute information from the interaction apparatus and outputting language data sent via a network, a data analysis part for parsing the input language data according to the attribute information by reference to the attributive phrase data groups in the attributive phrase data holding part, a data conversion part for determining whether the obtained analyzed phrase data are a conversion object or not, and converting analyzed phrase data found to be a conversion object according to conversion directions to create converted language data, and a data communication part for sending the converted language data to a networked interaction apparatus. The language data and video data output from one interaction apparatus are thus converted. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an interactive apparatus connected to a receiving-side interactive apparatus via a network, and by converting data to be transmitted, a user of the receiving-side interactive apparatus can use the transmission source from the transmitted data. The present invention relates to an interactive apparatus that prevents estimation of a user's identity and enables confidentiality protection of a transmission source user.

  Video telephones and electronic bulletin boards are known as interactive devices that are connected via a network and use multi-media data such as language data and video data to perform data communication. In the interactive apparatus as described above, prosodic information is used when text data is converted into speech data. This prosodic information is prosodic information estimated only from text data, and speech data with high naturalness cannot be synthesized by conversion to speech data using the prosodic information.

Therefore, as a countermeasure against the above problem, in the dialogue apparatus described in Patent Document 1, in addition to prosodic information estimated from text data, personality information and prosodic information estimated by analyzing actual speech data Is used as additional prosodic information and is used when synthesizing speech data, thereby realizing speech data synthesis with higher naturalness.
JP-A-10-320170

  By the way, the dialogue apparatus as described above has the following problems to be solved. That is, when language data and own video data are output from one interactive device to the other interactive device via the network, the language data transmitted from the transmitting interactive device is output as it is in the receiving interactive device. In addition, as the self video data, the video data transmitted from the transmission source interactive device is reproduced as it is. For this reason, the user of the dialog device on the receiving side can easily estimate the identity of the user of the dialog device of the transmission source such as gender, age, and locality from the data transmitted from the dialog device of the transmission source. there were. Therefore, if the user of the dialog device on the receiving side is a malicious person, the user who is presumed from the received data will be slandered, or their There was a risk that malicious acts such as streaming hoax information of users would be done. Therefore, there has been a demand for the development of an interactive device that can protect the confidentiality of users.

  Therefore, the present invention conceals the identity of the user of the transmission source by converting the language data and the video data when performing data transmission in the interactive apparatus connected via the network. An object of the present invention is to provide an interactive apparatus capable of improving confidentiality protection.

The present invention adopts the following configuration in order to solve the above points.
<Configuration 1>

  The present invention relates to an attribute word / phrase data holding unit that holds attribute word / phrase data groups associated with common word information for each attribute information for classifying words in an interactive device connected via a network, and for each common word information. A common word / attribute correspondence table in which attribute information that can be held is associated; input means for inputting language data to be transmitted and attribute information for instructing conversion; and analyzing the input language data A data analysis unit that divides into a plurality of analysis word / phrase data and refers to the attribute word / phrase data holding unit to search for common word information corresponding to each analysis word / phrase data, and refers to the common word / attribute correspondence table When each common word information holds the same attribute information as the conversion instruction attribute information, the attribute word / phrase data holding unit is searched based on the attribute information, and each common word is searched. A data conversion unit that takes in attribute phrase data synonymous with the information and converts each of the analysis phrase data into the attribute phrase data to generate conversion language data; and a dialog on the relative side of the conversion language data via the network And a data communication unit for transmitting to the apparatus.

<Configuration 2>
The interactive apparatus according to Configuration 1, wherein a video data holding unit that holds video data associated with attribute information, and corresponding video data is selected from the video data holding unit based on conversion instruction attribute information, Video data conversion means for converting video data into the video data;
It is characterized by providing.

<Configuration 3>
A method for converting language data of an interactive device, an attribute word / phrase data holding unit for holding an attribute word / phrase data group associated with common word information for each attribute information for classifying words / phrases in an interactive device connected via a network A common word / attribute correspondence table in which attribute information that can be held in each of the common word information is associated, and a step of inputting language data to be transmitted by the input means and attribute information for instructing conversion, Analyzing the linguistic data and dividing it into a plurality of pieces of analysis word / phrase data, searching the common word information corresponding to each piece of analysis word / phrase data with reference to the attribute word / phrase data holding unit, Referring to the attribute correspondence table, determining whether each of the common word information holds the same attribute information as the conversion instruction attribute information, and the common word information is the same Holding sex information, searching the attribute phrase data holding unit based on the attribute information, capturing attribute phrase data synonymous with the common word information, and analyzing the phrase data as the attribute phrase A language data conversion method for an interactive device, comprising: converting the data into data to generate converted language data; and transmitting the converted language data to the other interactive device via the network.

  According to the present invention, the attribute word / phrase data holding unit holding the attribute word / phrase data group associated with the common word information for each attribute information, and the common word When an attribute correspondence table is provided and language data and its attribute information are received from the input terminal, text data converted from the received language data and attribute information are transferred, and the input text data is converted to text data attribute information. Based on the attribute word / phrase data holding unit, the analysis word / phrase data group is referred to, the analysis word / phrase data is parsed, and the common word information corresponding to the analysis word / phrase data is taken in. Conversion text data is generated based on the common word information and conversion instruction attribute information associated with the data, and the conversion text data is By transmitting the converted text data to language data and outputting the language data, for example, the language data input by the woman is the same language data input by the man. Therefore, it is possible to conceal the identity of the user at the transmission source estimated from the language data received by the interactive apparatus on the receiving side, and to improve the confidentiality protection of the user.

  Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS.

  FIG. 1 is a block diagram of an interactive apparatus to which an embodiment of the present invention is applied. This interactive device 2 is connected to another interactive device 10 via a communication network 9. The interactive device 2 includes an input terminal such as a keyboard and an output terminal such as a monitor. The interactive device 2 further includes an input / output control unit 1, a data analysis unit 3, a data conversion unit 4, a language information holding unit 5, and a data communication unit 8.

  The input / output control unit 1 receives the text data from an input terminal such as a keyboard and transfers the data to the data analysis unit 3. The input control unit 1 A monitors the text data received via the network. And an output control unit 1B that outputs to the output terminal. When the input of the text data is finished, the output control unit 1B controls the output terminal to display a screen for inputting attribute information of the data and a screen for instructing attribute information to be converted.

  FIG. 2 is an example of attribute information. That is, the attribute information includes attribute information indicating dialects such as “Tokyo dialect” and “Kansai dialect”, attribute information indicating age such as “young” and “old man”, and sex such as “male” and “female”. The attribute information to be displayed is displayed so as to be selectable.

  The language information holding unit 5 includes an attribute word / phrase data holding unit 6 and a common word / attribute correspondence table 7 in which common word information and attribute information are associated with each other. The attribute word / phrase data holding unit 6 holds an attribute word / phrase data group associated with common word information for each attribute information. FIG. 3 is an example of attribute phrase data having an attribute of “Kansai dialect” in the attribute phrase data group. Common word information such as “How is it” and “Good” is associated with the expressions “Kanasai dialect” attribute phrase data “Do not it is” and “Yes”, respectively.

  Further, the common word / attribute correspondence table 7 is associated with attribute information groups provided for each common word information. FIG. 4 is an example of attribute information for the common word information “good”, and attribute phrase data synonymous with “good” corresponding to each attribute information “Tokyo dialect”, “Kansai dialect”, and “old man”. This indicates that the attribute word / phrase data holding unit 6 holds it.

  When the data analysis unit 3 receives the text data from the input / output control unit 1, the data analysis unit 3 refers to the attribute phrase data information group of the attribute phrase data holding unit 6 based on the attribute information selected and input by the user, and Analysis is performed, and each word data and corresponding common word information obtained as a result of the analysis are transferred to the data converter 4. For example, the phrase data “Yes” and the common word information “Good” are transferred.

  The data conversion unit 4 searches the common word / attribute correspondence table 7 using the common word information corresponding to each phrase data obtained from the data analysis unit 3 as a search element, takes in the attribute information held in the common word information, and designates conversion It is determined whether or not the attribute information is present in the acquired attribute information, and if it exists, the attribute word / phrase data group divided into the attribute information of the conversion instruction in the attribute word / phrase data holding unit 6 is converted into the common word information. Corresponding attribute phrase data is extracted, converted text data is generated from the extracted attribute phrase data, and the converted text data is transferred to the data communication unit 8. If the attribute information specified for conversion does not exist in the captured attribute information, the phrase data is used as it is for conversion text data generation and transferred to the data communication unit 8 without being converted. In addition, what is not contained in attribute phrase data, such as a noun, shall be utilized for conversion text data generation, without converting as it is.

  When the data communication unit 8 receives the converted text data from the data conversion unit 4, the data communication unit 8 transmits the converted text data to the partner interactive device 10 via the network and receives the text data from the partner interactive device 10. Then, the received data is transferred to the input / output control unit 1.

Next, the operation of the interactive apparatus 2 configured as described above will be described below.
FIG. 5 is a flowchart of the operation of the interactive apparatus 2 according to the first embodiment.

  In the first embodiment, the text data to be input is “Today's good weather”, and “Tokyo dialect”, “female”, and “young” are selected and input as attribute information of the text data, and conversion instructions are given. As attribute information to be entered, “Tokyo dialect”, “male”, and “young” are entered. Therefore, the text data entered by “female” is the same text data entered by “male”. Is converted to

  First, when text data is input from an input terminal such as a keyboard and the text data attribute information and conversion instruction attribute information are input via the output terminal, the input / output control unit 1 receives the received data. The data is transferred to the data analysis unit 3 (step S1).

FIG. 6 is an operation diagram of data analysis of the dialogue apparatus 2.
When the data analysis unit 3 receives the text data such as “It ’s good weather today” transferred from the input / output control unit 1 and the attribute information of the text data, the data analysis unit 3 stores the text information of the language information holding unit 5 based on the attribute information. By referring to the attribute word / phrase data group of each attribute information such as “Tokyo dialect”, “female”, and “young” from the attribute word / phrase data holding unit 6, and searching for word / phrase data matching each word / phrase in the text, , The above text data is analyzed as “today”, “ha”, “good”, “weather”, “dawa”, and the analysis phrase data “today” “ha” “good” “weather” “dawa” and “good” ”And“ Dawa ”are acquired corresponding common word information“ good ”and“ is ”, respectively, and the acquired analysis phrase data and the common word information are transferred to the data conversion unit 4 (step S2). In addition, what does not correspond to attribute phrase data, such as a noun and a particle, is text-analyzed as a noun or a particle as it is.

FIG. 7 is an operation diagram of data conversion of the dialogue apparatus 2.
In the first embodiment, the conversion process of the analysis phrase data “daiwa” will be described.
The data conversion unit 4 accepts “today”, “ha”, “good”, “weather”, “dawa”, common word information “I”, and conversion instruction attribute information, which are analysis phrase data transferred from the data analysis unit 3. Then, conversion processing is performed for each analysis phrase data. Specifically, the data conversion unit 4 uses the common word information “I” associated with the analysis word / phrase data “DAWA” captured by the data analysis unit 3 as a search element, and the common word / attribute correspondence table 7. And the attribute information “Kansai dialect”, “old man”, and “male” held in the common language information “is” are captured.

  Then, it is determined whether or not the attribute information “Tokyo dialect”, “young”, and “male” of the conversion designation exists in the captured attribute information. In the first embodiment, since the attribute information “male” of conversion designation is present in the acquired attribute information, the data conversion unit 4 uses the attribute word / phrase data holding unit 6 to classify the attribute word / phrase into the attribute information “male” of the conversion instruction. The attribute word / phrase data “DANA” corresponding to the common word information “DA” is extracted from the data group, and “DAWA” is converted into the extracted attribute word data “DANA” (step S3). If the conversion designation attribute information is not found in the retrieved attribute information by searching the common word / attribute correspondence table 7 based on the common word information, the data conversion unit 4 No conversion is performed.

  After the conversion of all the analysis phrase data to be converted, the data conversion unit 4 generates the converted text data “Today is good weather” and outputs it to the data communication unit 8 (step S4).

  When the data communication unit 8 receives the converted text data “It's good weather today” from the data conversion unit 4, the data communication unit 8 sends the converted text data to the input / output control unit of another interactive apparatus 10 connected via the network. Transmit (step S5).

  The input / output control unit of the other interactive device 10 displays the converted text data “Today's good weather” transmitted from the data communication unit 8 on the screen via its own output terminal (step S6). .

  As described in the first embodiment, the attribute word / phrase data holding unit that holds the attribute word / phrase data group associated with the common word information for each attribute information and the attribute information group that the user owns for each common word information are associated with each other. When the text data and its attribute information are received from the input terminal, the received text data and attribute information are transferred. Then, based on the attribute information of the text data, the input text data is referred to the attribute word / phrase data group from the attribute word / phrase data holding unit, parsed into the analysis word / phrase data, and common word information corresponding to the analysis word / phrase data The converted text data is generated based on the common word information and the conversion instruction attribute information associated with the analyzed phrase data, and the converted text data is connected via the network. By transmitting to another interactive device, for example, text data input by a woman can be converted into the same text data input by a male, and therefore estimated from the text data received by the receiving interactive device It is possible to conceal the identity of the user at the transmission source and to improve the confidentiality protection of the user.

FIG. 8 is a block diagram of the interactive apparatus according to the second embodiment.
When the speech data is received by the input / output control unit 1 of the interactive device of the first embodiment shown in FIG. 1 from the input terminal such as a microphone, the interactive device 2 of the second embodiment converts the voice data into text data. A voice recognition unit 1C for transferring to the data analysis unit 3, a voice conversion unit 1D for converting converted text data received via the network into voice data, and transferring the voice data to an output terminal such as a speaker, and input / output control The unit 1 specifies the prosodic information group to be used by the other interactive device 10 connected via the network when converting the attribute information of the conversion designation selected and input from the output terminal screen when the converted text data received is converted into speech. Is transferred to the data communication unit 8 as converted prosody specific information.

  When another interactive device 10 via the network receives the converted prosody specific information holding unit 10A that holds the converted prosody specific information transmitted from the interactive device 2 and the converted text data, the converted text data is based on the prosodic information. A prosody information holding unit 10B that holds a group of prosodic information indicating the level of pronunciation and strength associated with each attribute word / phrase data for each attribute information indicating a dialect for conversion into speech data is provided. As the prosody information group held by the above-mentioned prosody information holding unit 10B, “Tokyo dialect prosody information group”, “Kyushu dialect prosody information group”, and the like can be considered. As a result, when the converted text data is received, for example, when the converted prosody specific information is “Kyushu dialect”, it is possible to convert it into voice data that has been adjusted to prosody based on the characteristics of the Kyushu dialect.

  Similarly, the input / output control unit 1 also includes a conversion prosody specifying information holding unit 1E that holds the converted prosody specifying information transferred from the data communication unit 8. Further, the prosodic information group as held by the prosodic information holding unit 10B is held in the language information holding unit 5, and when the input / output control unit 1 accepts the converted text data, it has been transferred and held in advance. Based on the converted prosody specifying information, the prosodic information group is referred to from the language information holding unit 5, the prosodic information group to be used is specified, the converted text data is converted into speech data based on the specified prosodic information, and converted. The voice data is output through an output terminal provided by itself.

  When the data communication unit 8 receives the converted text data from the data conversion unit 4 or the conversion prosody identification information from the input / output control unit 1, the data communication unit 8 transmits the received data to the other-side dialog device 10 via the network, and When the converted text data or the converted prosody specific information is received from the other-side dialog device 10, the received data is transferred to the input / output control unit 1. The other data analysis unit 3 and data conversion unit 4 are the same as those in the first embodiment.

Next, the operation of the interactive apparatus having the above configuration will be described below.
FIG. 9 is a flowchart of the operation of the interactive apparatus according to the second embodiment.

  In the second embodiment, the voice data to be input is “Today is good weather”, and “Tokyo dialect”, “female”, and “young” are selected and input as attribute information of the voice data, and conversion instructions are given. As attribute information to be input, “Tokyo dialect”, “male”, and “young” are input. Therefore, the audio data input by “female” is the same audio data input by “male”. Is converted to

  First, when voice data is input from an input terminal such as a microphone and the attribute information of the voice data and the attribute information of the conversion instruction are input via the output terminal, the input / output control unit 1 performs voice recognition 1C. The voice data is converted into text data, and the converted text data is transferred to the data analysis unit 3 (steps S1 and S2). Also, the conversion designation attribute information received at this time is also transferred to the data communication unit 8 as conversion prosody identification information, and is transmitted to the data communication unit of another interactive apparatus 10 via the network. The data communication unit of the dialogue apparatus 10 transfers the conversion prosody specific information to the input / output control unit provided therein, and the input / output control unit uses the prosody adjustment processing when converting the received converted text into speech data. The conversion prosody specific information is held.

  When the data analysis unit 3 receives the text data “Today is a good weather” and the attribute information of the text data transferred from the input / output control unit 1, the attribute phrase data of the language information holding unit 5 is received based on the attribute information. By referring to the attribute word / phrase data group of each attribute information such as “Tokyo dialect”, “female”, and “young” corresponding from the holding unit 6 and searching for the phrase data matching each word of the text, the above-mentioned The text data is analyzed as “Today”, “Ha”, “Good”, “Weather”, “Dawa”, and the analysis phrase data and the corresponding common word information “Good”, “I” are acquired. Then, the obtained analysis phrase data is transferred to the data converter 4 (step S3). As in the first embodiment, texts that are not included in attribute word / phrase data such as nouns and particles are analyzed as texts as nouns or particles as they are.

  When the data conversion unit 4 receives the analysis word / phrase data “today”, “ha”, “good”, “weather”, “dawa” and the common word information “I” and the conversion instruction attribute information transferred from the data analysis unit 3, A conversion process is performed for each analysis phrase data. In the second embodiment, the conversion process of the analysis phrase data “daiwa” will be described.

  The data conversion unit 4 searches the common word / attribute correspondence table 7 using the common language information “D” associated with the analysis phrase data “Da” taken in by the data analysis unit 3 as a search element. The attribute information “Kansai dialect”, “old man”, and “male” stored in the word information “Da” are captured.

  Then, it is determined whether or not the attribute information “Tokyo dialect”, “young”, and “male” of the conversion designation exists in the captured attribute information. Since the attribute information “male” of the conversion designation is present in the attribute information acquired this time, the data conversion unit 4 continues the attribute word / phrase data group classified into the attribute information “male” of the conversion instruction in the attribute word / phrase data holding unit 6. The attribute word / phrase data “Dana” corresponding to the common word information “Da” is extracted from the list, and the extracted attribute word data “Dana” and “Dada” are converted (step S4). If the conversion designation attribute information is not found in the retrieved attribute information by searching the common word / attribute correspondence table 7 based on the common word information, the data conversion unit 4 No conversion is performed.

  After the conversion of all the analysis phrase data to be converted, the data conversion unit 4 generates the converted text data “Today's good weather” and outputs it to the data communication unit 8 (step S5).

  When the data communication unit 8 receives the converted text data “Today's good weather” from the data conversion unit 4, the data communication unit 8 sends the converted text data “Today” to the data communication unit of another interactive apparatus 10 connected via the network. "It's good weather" is transmitted (step S6).

  The data communication unit of the dialogue apparatus 10 transfers the received converted text data “Today's good weather” to the input / output control unit provided therein, and the input / output control unit converts the converted text data “Today has good weather”. When “NA” is accepted, the conversion prosody indicating the dialect of the conversion prosody specification information “Tokyo dialect”, “young”, and “male” is referred to by referring to the conversion prosody specification information that has been transmitted from the dialogue apparatus 2 and held in advance. Capture specific information "Tokyo dialect".

  Then, the input / output control unit refers to the “Tokyo dialect prosody information group” indicating the prosody information of the corresponding dialect from the prosody information holding unit 10B based on the captured conversion prosody specific information “Tokyo dialect”. In addition, from the “Tokyo dialect information group”, the prosody information such as pronunciation and strength associated with the phrase data that matches the phrase of the text of the converted text data “Today's good weather” is specified, and each phrase of the text Prosodic information such as pronunciation and strength peculiar to Japanese dialect is acquired. Subsequently, based on the prosodic information such as pronunciation and strength corresponding to the acquired “Tokyo dialect” word data, the input / output control unit adjusts the prosody of the converted text data to the Tokyo dialect and converts it into speech data. . Then, the converted audio data is output via an output terminal provided in itself such as a speaker. (Steps S7 and S8).

  As described in the second embodiment, the attribute word / phrase data holding unit that holds the attribute word / phrase data group associated with the common word information for each attribute information and the attribute information group that the self has for each common word information are associated with each other. A common word / attribute correspondence table, and the prosody that shows the level of pronunciation and strength associated with each attribute word / phrase data according to the attribute information indicating the dialect to other interactive devices connected via the network When the language data and its attribute information are received from the input terminal, the received language data is converted to text, and the obtained text data and its attribute information are transferred.

  Then, based on the attribute information of the text data, the text data obtained by input and text conversion is referred to the attribute phrase data group from the attribute phrase data holding unit, parsed into the analysis phrase data, and the analysis phrase data Based on the common word information and the conversion instruction attribute information associated with the analysis word / phrase data, the conversion text data is generated, and the conversion text data is obtained. When transmitted to another interactive device connected via the network and the other interactive device accepts the converted text data, the prosodic information group based on the converted prosodic information is referred to the prosodic information holding unit, and the specified prosodic information For example, audio data input by a woman is converted to audio data based on It is possible to convert, to hide the source of the user's identity the recipient of the interactive device is estimated from the language data that accepts, it is possible to improve the user's confidentiality protection.

  The present invention is not limited to the above embodiments. The first and second embodiments have been described with reference to the example of the interactive device that converts language data. For example, the interactive device has a video data holding unit that holds video data newly associated with attribute information indicating sex. By selecting video data based on the conversion instruction attribute information from the video data holding unit and providing video data conversion means for converting the video data from self-video data, conversion of not only language data but also video data is possible. This makes it possible to conceal the user's identity estimated from the language data and the video data, and to improve the confidentiality protection of the user.

  In addition, the language information holding unit 5 referred to for data analysis and conversion in the interactive devices of the first and second embodiments can be realized by using an expression format such as RDF or ontology.

  Further, in the interactive devices of the first and second embodiments, what is not included in the attribute word / phrase data such as nouns is used for generating converted text data without being converted as it is. In the attribute information “young people” in the attribute phrase data “part-time”, the synonymous attribute phrase data may be different depending on the attribute information so as to be associated with the attribute phrase data “byte”. It is also possible to realize a configuration in which attribute word / phrase data such as nouns is included in the language information holding unit 5 so that nouns can be converted.

It is a block block diagram of the dialogue apparatus of Example 1. It is a figure which shows attribute information. It is a figure which shows attribute phrase data. It is a figure which shows a common word and attribute correspondence table. 3 is an operation flowchart of the interactive apparatus according to the first embodiment. It is an operation | movement figure of the data analysis process of the dialogue apparatus of Example 1. It is an operation | movement diagram of the data conversion process of the dialogue apparatus of Example 1. It is a block block diagram of the dialogue apparatus of Example 2. 6 is an operation flowchart of the interactive apparatus according to the second embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Input / output control part 1A Input control part 1B Output control part 1C Speech recognition part 1D Speech conversion part 1E Conversion prosody specific information holding part 2 Dialogue device 3 Data analysis part 4 Data conversion part 5 Language information holding part 6 Attribute word phrase data holding part 7 Common Language / Attribute Correspondence Table 8 Data Communication Unit 9 Communication Network 10 Dialogue Device 10A Conversion Prosody Specific Information Holding Unit 10B Prosody Information Holding Unit

Claims (3)

In an interactive device connected via a network,
An attribute word / phrase data holding unit holding an attribute word / phrase data group associated with common word information according to attribute information for classifying words;
A common word / attribute correspondence table in which attribute information that can be held for each common word information is associated;
Input means for inputting language data to be transmitted and attribute information for instructing conversion;
A data analysis unit that analyzes the input language data and divides it into a plurality of analysis word / phrase data and searches for common word information corresponding to each analysis word / phrase data with reference to the attribute word / phrase data holding unit;
If the common word information holds the same attribute information as the conversion instruction attribute information with reference to the common word / attribute correspondence table, the attribute word / phrase data holding unit is searched based on the attribute information. A data conversion unit that captures attribute phrase data having the same meaning as each of the common word information, converts each analysis phrase data into the attribute phrase data, and generates conversion language data;
A data communication unit for transmitting the converted language data to the interactive apparatus on the opposite side via the network;
An interactive apparatus characterized by including: The interactive apparatus according to claim 1,
A video data holding unit that holds video data associated with the attribute information;
Video data conversion means for selecting the corresponding video data based on the conversion instruction attribute information from the video data holding unit and converting the self-video data into the video data;
An interactive apparatus comprising: An attribute word / phrase data holding unit that holds attribute word / phrase data groups associated with common word information for each attribute information to classify words / phrases in an interactive device connected via a network, and can be held in each common word information A common word / attribute correspondence table with attribute information is provided,
A step of inputting language data to be transmitted by the input means and attribute information for instructing conversion;
Analyzing input language data and dividing it into a plurality of analysis phrase data;
Referring to the attribute phrase data holding unit, searching for common word information corresponding to each analysis phrase data;
Determining whether each of the common word information holds the same attribute information as the conversion instruction attribute information with reference to the common word / attribute correspondence table;
When the common word information holds the same attribute information, the attribute word phrase data holding unit is searched based on the attribute information, and the attribute word phrase data having the same meaning as the common word information is acquired.
Converting the analysis phrase data into the attribute phrase data to generate conversion language data;
Transmitting the converted language data to the other dialog device via the network. A method for converting language data of the dialog device.

Images