JP2022009228A - Document generation device, document generation method, and program for document generation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a document generation device, a document generation method, and a program that allows a user to easily input.

SOLUTION: A document generation device includes: a document generation unit 22f that generates a document according to a result of analyzing text data converted from an audio signal spoken by a speaker; a keyword extraction unit 22e that extracts a keyword from the text data according to the analysis result; an additional information generation unit 22g that generates, based on at least one of the document and the keyword, additional information to be added to the document; and a message generation unit 23 that generates a message according to the document and the additional information.

SELECTED DRAWING: Figure 7

COPYRIGHT: (C)2022,JPO&INPIT

Description

The present application belongs to the technical fields of a document generator, a document generator, and a program for a document generator.

A device for sending an email while driving a vehicle, which is an example of a moving body, has been developed. For example, in the following Patent Document 1, the mail fixed phrase received from the mobile communication terminal is stored in the mail fixed sentence storage unit, and in response to the mail transmission request, the mail fixed sentence stored in the mail fixed sentence storage unit is stored. An in-vehicle terminal that selects one of the above and sends it to the mobile communication terminal, and an in-vehicle terminal that sends the generated email fixed phrase to the in-vehicle terminal, and also receives the email fixed phrase sent from the in-vehicle terminal, and this received email. A mobile communication terminal that generates an e-mail to be sent by an e-mail sending unit based on a fixed phrase and sends it to a e-mail server, and an e-mail sending system using the e-mail transmission system are disclosed.

Japanese Unexamined Patent Publication No. 2006-333777

However, in the technique described in Patent Document 1, it is necessary to press a button to select an e-mail to be sent while driving, and it cannot be said that it is easy for the user to input. Also, it was necessary to select, search for, and enter information related to the content to be sent.

Therefore, the present application has been made in view of each of the above problems, and one example of the problem is to provide a document generation device or the like that is easy for a user to input.

In order to solve the above problems, the invention according to claim 1 has a document generation unit that generates a document according to the result of analyzing text data converted from a voice signal spoken by the speaker, and the result of the analysis. An extraction unit that extracts keywords from the text data, an additional information generation unit that generates additional information to be added to the document based on at least one of the document and the keyword, and the document and the additional information. It is characterized by having a message generation unit that generates a message according to the above.

Further, the invention according to claim 6 is the document generation step of generating a document according to the result of analyzing the text data converted from the voice signal spoken by the speaker, and the invention according to the analysis result. An extraction step for extracting keywords from text data, an additional information generation step for generating additional information to be added to the document based on at least one of the document and the keyword, and a message according to the document and the additional information. It is characterized by including a message generation step for generating a message.

It is a block diagram which shows an example of the outline structure of the document generation apparatus which concerns on embodiment. It is a block diagram which shows an example of the outline structure of the document generation apparatus which concerns on Example. It is a schematic diagram which shows an example of the address database constructed in the storage part of FIG. It is a schematic diagram which shows an example of the keyword database constructed in the storage part of FIG. It is a schematic diagram which shows an example of the additional information database built in the storage part of FIG. It is a block diagram which shows an example of the functional block of the document generation apparatus of FIG. It is a block diagram which shows an example of the functional block of the sentence analysis part of FIG. It is a flowchart which shows an example of the operation of the document generation apparatus which concerns on Example. It is a schematic diagram which shows the content uttered by the speaker and an example of the generated message. It is a flowchart which shows an example of the subroutine of analysis of the morphological analysis result of FIG. It is a flowchart which shows an example of the subroutine of the message generation of FIG. It is a schematic diagram which shows the content uttered by the speaker and an example of the generated message. It is a schematic diagram which shows the content uttered by the speaker and an example of the generated message. It is a schematic diagram which shows the content uttered by the speaker and an example of the generated message.

A mode for carrying out the present application will be described with reference to FIG. Note that FIG. 1 is a block diagram showing an example of an outline configuration of the document generator according to the embodiment.

As shown in FIG. 1, the document generation device 10 includes a document generation unit 10a, an extraction unit 10b, an additional information generation unit 10c, and a message generation unit 10d that generates a message according to a document and additional information. It is configured in preparation.

In this configuration, the document generation unit 10a generates a document according to the result of analyzing the text data converted from the voice signal spoken by the speaker.

Here, as an example of a speaker, a user of a moving body such as a vehicle, a motorcycle, an aircraft, or a ship can be mentioned.

In addition, as an example of the conversion process from a voice signal to text data, there is a voice recognition process for extracting what is being spoken as character data, and as an example of recognition technology, a statistical method, a dynamic time expansion / contraction method, and a hidden method can be mentioned. Markov model and the like can be mentioned.

Further, as an example of analysis of text data, there is a technique of natural language processing such as morphological analysis. In addition, as an example of the result of analyzing text data, the result of decomposing a document into morphemes (which may be words or phrases), which is the smallest unit having meaning in a language, by morphological analysis, etc., and the result of determining the part of speech of morphemes, etc. Can be mentioned.

In addition, as an example of a document, there are a fixed phrase "delayed" for the keyword "delayed", a fixed phrase "delayed" for the keyword "late", a fixed phrase "pass now" for the keyword "current location", and the keyword "now".

In this configuration, the extraction unit 10b extracts keywords from the text data according to the analysis result.

Here, as an example of extracting keywords from text data according to the analysis result, extraction and "display" of keywords (an example of output keywords) meaning transmission such as "send", "send", and "contact". , Extraction of keywords (an example of output keywords) that mean display such as "Show", Extraction of keywords that mean output destinations such as mail destinations such as person name and company name, and display destinations such as "Display", " Extraction of keywords such as "delayed" and "late" for extracting the fixed phrase "delayed", extraction of keywords for requesting additional information, and the like.

The additional information generation unit 10c generates additional information to be added to the document based on at least one of the document and the keyword.

Here, as an example of additional information, map information (latitude, longitude) for the keyword "current location", road information for the keyword "congestion", and the current location or route for the keywords related to weather such as "rain" and "snow". Weather information, arrival time information for the fixed phrase "I'll be late", map information (latitude, longitude) for the fixed phrase "I'm passing here now", etc.

The message generation unit 10d generates a message according to the document and additional information.

Here, as an example of a message, when there is a keyword "current location", a message including a map of the current location added to the document, a message including a document "map display" indicating a link of the map of the current location, and a keyword "traffic jam" are used. If there is a document that includes the document "○○ to ○ km traffic jam", or if there is a keyword "wait", there is a document such as "please wait at the east exit of ○○ station".

As described above, according to the operation of the document generation device 10 according to the embodiment, it is possible to complement the words spoken by the speaker and generate a natural message.

[1. Document generator configuration and function overview]
(1.1 Overview of document generator configuration and functions)

Next, specific examples corresponding to the above-described embodiments will be described with reference to FIGS. 2 to 5. The examples described below are examples when the present application is applied to a document generator.

FIG. 2 is a block diagram showing an example of an outline configuration of a document generator according to an embodiment. FIG. 3 is a schematic diagram showing an example of an address database constructed in the storage unit. FIG. 4 is a schematic diagram showing an example of a keyword database constructed in the storage unit. FIG. 5 is a schematic diagram showing an example of an additional information database constructed in the storage unit.

As shown in FIG. 2, the document generation device 10 according to the embodiment includes an input unit 11 that accepts input from a user, a storage unit 12 that stores a program or the like for controlling the document generation device 10, and a document generation device. Communication unit 13 that transmits and receives to and from the outside of 10, sensor unit 14 that detects the state of the moving object used by the user, output unit 15 that outputs the processing result, and system control unit that controls the document generation device 10. A 16 and an input / output interface unit 17 are provided. The system control unit 16 and the input / output interface unit 17 are connected to each other via the system bus 18.

The input unit 11 includes a microphone that converts the voice spoken by the speaker into an electric signal, a touch pad having a position input function, and the like. The position input function portion is composed of elements that realize, for example, a capacitance method, a resistance film method, and the like. Then, the input unit 11 acquires the position information of the surface of the input unit 11 that the user's finger or the like touches or is close to by the position input function in pixel units with a predetermined resolution. The input unit 11 may be a touch panel having a display function and a position input function.

The input unit 11 receives voice from the user and receives input such as a destination from the user.

The storage unit 12 is composed of, for example, a silicon disk drive or the like. The storage unit 12 stores various programs and the like for controlling the document generation device 10. Examples of various programs include an operating system and a program for sending an e-mail. The various programs may be acquired via a network such as a wireless communication network, or may be recorded on a recording medium such as a CD (Compact Disc) or DVD (Digital Versatile Disc) to drive the drive device. It may be read via.

Further, the storage unit 12 is constructed with an address database for storing e-mail address information, a keyword database for storing keywords and fixed phrases, an additional information database for storing additional information to be added to a document, and the like. Further, a dictionary database for performing morphological analysis is constructed in the storage unit 12. Further, a map database is constructed in the storage unit 12.

Further, the storage unit 12 has an output database that stores keywords related to output such as transmission and display. For example, keywords such as "contact", "send", "mail", and "display" are stored in the output database.

As shown in FIG. 3, the address database stores e-mail addresses in association with personal names, company names, place names, and the like. For address input to the address database, an address book such as a smartphone can be acquired by communication, an address can be directly input to an in-vehicle system including a document generator 10, or a communication unit 13 can be input from an address book on the cloud of the network. Or get through.

As shown in FIG. 4, a fixed phrase is stored in the keyword database in association with the keyword. A fixed phrase number is assigned to each fixed phrase.

Further, in FIG. 4, in the case of a keyword such as "wait" in which a message accompanied by a place is input, a fixed phrase indicating a place to enter a place word such as "*" is prepared.

The fixed phrase input to the keyword database may be registered from an external device such as a personal computer or a smartphone.

Further, when the morphological analysis of the speech recognition result is performed and the result is treated as a keyword, the combination of the stem and the conjugation of the verb may be treated as a keyword. For example, in the case of "delayed", the keyword is a combination of stem: "slow" and flexion: lower one-row conjugation: "re, re, re, re, re, re, rero". Further, a combination of morphemes having different parts of speech may be treated as a keyword. For example, for "wait", "wait": an independent verb, "te": a particle, a connecting particle, and "te": a non-independent verb. Also, for example, for "te", "i": an independent verb, "te": a particle, and a connecting particle.

As shown in FIG. 5, the additional information database stores additional information in association with a keyword or a fixed phrase. In addition, additional information may be associated by a combination of a keyword and a fixed phrase.

The additional information may be registered in the additional information database from an external device such as a personal computer or a smartphone. Further, the additional information is generated by using the information from the sensor unit 14, the time information, the map database, and the like. Further, in the additional information database, sentences / messages related to additional information may be dynamically registered as fixed phrases. Further, in the additional information database, it may be statically registered in advance as a sentence / message that handles additional information. Further, there may be a plurality of additional information databases depending on the type of additional information.

The communication unit 13 is configured to connect to a network such as the Internet by wireless communication and control the communication state with the outside.

The sensor unit 14 has a GPS (Global Positioning System) sensor, receives navigation radio waves from GPS satellites in satellite orbit, and detects the position, altitude, and the like of a moving object in which the document generator 10 is installed. Further, the sensor unit 14 is connected to various sensors of the moving body (for example, a sensor that captures a change in the liquid level due to tilt, a vehicle speed pulse, an acceleration sensor or a gyro sensor), and the speed, acceleration, tilt, mileage, and mileage of the moving body. Acquires moving object information such as engine speed. The sensor unit 14 is an example of a current location acquisition unit that acquires the current location.

The output unit 15 is composed of a speaker that outputs sound, a liquid crystal display element that displays a generated document, an EL (Electro Luminescence) element, and the like. When the display function of the output unit 15 is a head-up display, the output unit 15 has a projection means and a screen means (windshield, lens unit of eyeglasses, etc.) for projecting an image projected from the projection means.

The system control unit 16 is, for example, a CPU (Central Processing Unit) 16a and an RO.
It has an M (Read Only Memory) 16b and a RAM (Random Access Memory) 16c. The system control unit 16 reads and executes various programs stored in the ROM 16b, the RAM 16c, and the storage unit 12 by the CPU 16a. The system control unit 16 performs voice recognition, analyzes a document, generates a message, and calculates a movement route of a moving object based on information on a current location and a received destination.

Further, the document generation device 10 has an internal clock (not shown).

(1.2 Functional block of document generator)
Next, the functional block of the document generation device 10 will be described with reference to FIGS. 6 and 7.
FIG. 6 is a block diagram showing an example of a functional block of the document generation device 10. FIG. 7 is a block diagram showing an example of a functional block of the document analysis unit 22.

As shown in FIG. 6, the document generation device 10 outputs to the voice recognition unit 21 that converts the voice data acquired by the input unit 11 into text data, the document analysis unit 22 that analyzes the text data, and the output unit 15. It has a message generation unit 23 for generating a message, and a message generation unit 23. The voice recognition unit 21, the document analysis unit 22, and the message generation unit 23 are realized by the system control unit 16.

The message generation unit 23 (an example of the message generation unit 10d) creates a fixed phrase according to the analysis result of the document analysis unit 22. The message generation unit 23 adds a text analyzed by morphological analysis to the selected fixed phrase, or adds a fixed phrase including additional information.

The document analysis unit 22 performs morphological analysis of text data with reference to a dictionary database, and performs address analysis, keyword analysis for selecting fixed phrases, analysis for additional information, and the like on the analysis results. The document analysis unit 22 is an example of the document generation unit 10a, the extraction unit 10b, and the additional information generation unit 10c.

As shown in FIG. 7, the document analysis unit 22 includes a morphological analysis unit 22a, an output keyword extraction unit 22b, an output destination extraction unit 22c, an output determination unit 22d, a keyword extraction unit 22e, and a document generation unit 22f. , And an additional information generation unit 22g. The keyword extraction unit 22e, the document generation unit 22f, and the additional information generation unit 22g form an analysis unit 25.

The morphological analysis unit 22a performs morphological analysis on the text data converted by the voice recognition unit 21, decomposes it into morphemes, and obtains a part of speech and the like.

The output keyword extraction unit 22b (an example of the extraction unit 10b) has an output database (storage unit 12). The output keyword extraction unit 22b refers to the output database and extracts words meaning output such as transmission and display from the decomposed morphemes.

The output destination extraction unit 22c (an example of the extraction unit 10b) has an address database (storage unit 12). The output destination extraction unit 22c refers to the address database and extracts a person name, a company name, a place name, etc. from the decomposed morphemes. Next, the output destination extraction unit 22c refers to the address database and extracts the e-mail address corresponding to the person name, company name, place name, and the like.

The output determination unit 22d determines whether or not to output the generated message based on the output keyword extraction unit 22b and the output destination extraction unit 22c. For example, the output determination unit 22d determines that a message is output when the word meaning output and the destination e-mail address can be extracted.

The analysis unit 25 includes a keyword extraction unit 22e, a document generation unit 22f, an additional information generation unit 22g, a keyword database (storage unit 12), and an additional information database (storage unit 12).

The keyword extraction unit 22e (an example of the extraction unit 10b) refers to the keyword database and extracts keywords from the decomposed morphemes. For example, the keyword extraction unit 22e extracts keywords corresponding to fixed phrases. Further, the keyword extraction unit 22e may extract a keyword from the decomposed morphemes by referring to the additional information database. For example, the keyword extraction unit 22e extracts the keyword corresponding to the additional information.

The document generation unit 22f (an example of the document generation unit 10a) refers to the keyword database, extracts a fixed phrase corresponding to the keyword extracted by the keyword extraction unit 22e, and generates a document. The document generation unit 22f may generate a document by combining morphemes from the decomposed morphemes.

The additional information generation unit 22g (an example of the additional information generation unit 10c) refers to the additional information database, extracts the additional information corresponding to the keyword extracted by the keyword extraction unit 22e, and generates the additional information. Further, the additional information generation unit 22g extracts the additional information corresponding to the document generated by the document generation unit 22f, and generates the additional information. Further, the additional information generation unit 22g adds the current position as additional information based on the sensor unit 14, calculates the arrival time, and adds the arrival time as additional information.

The message generation unit 23 generates a message based on the result of the output destination extraction unit 22c, the result of the document generation unit 22f, and the result of the additional information generation unit 22g.

The output unit 15 outputs the message generated by the message generation unit 23 to the output destination based on the determination result of the output determination unit 22d.

[2. Operation of document generator]
(2.1 Operation example of document generator)
The operation of the document generator according to the embodiment will be described with reference to FIGS. 8 to 14.

FIG. 8 is a flowchart showing an example of the operation of the document generation device 10 according to the embodiment. FIG. 9 is a schematic diagram showing an example of the content spoken by the speaker and the generated message. FIG. 10 is a flowchart showing an example of a subroutine for analyzing the morphological analysis result of FIG. FIG. 11 is a flowchart showing an example of the subroutine for generating the message of FIG. 12 to 14 are schematic diagrams showing an example of the content spoken by the speaker and the generated message.

As shown in FIG. 8, the document generation device 10 accepts voice input (step S1). Specifically, the input unit 11 receives the voice spoken by the user (speaker) using the mobile body.

The document generation device 10 may receive the input of the destination, acquire the current location from the sensor unit 14, and obtain the route to the destination.

Next, the document generation device 10 converts the voice signal into text data by voice recognition (step S2). Specifically, the system control unit 16 (voice recognition unit 21) converts the voice signal into text data. As shown in FIG. 9, the system control unit 16 asks for the voice recognition result: "contact Tanaka with the current location".

Next, the document generator 10 morphologically analyzes the text data (step S3). Specifically, the system control unit 16 (morphological analysis unit 22a) analyzes the text data into morphemes and decomposes them into morphemes. As shown in FIG. 9, as a result of the morphological analysis, the system control unit 16 "tanaka", "ni", "current location", "", "contact", "", "to", "te". Find each morpheme like.

Next, the document generator 10 analyzes the morphological analysis result (step S4). Specifically, the system control unit 16 (output keyword extraction unit 22b, output destination extraction unit 22c, output determination unit 22d, analysis unit 25) analyzes the morphological analysis result of the morphological analysis unit 22a. The analysis of the morphological analysis result will be described in detail in the subroutine of the analysis of the morphological analysis result.

Next, the document generation device 10 determines whether or not to generate a message (step S5). Specifically, the system control unit 16 determines whether or not to generate a message based on the message generation flag as the return of the subroutine for analyzing the morphological analysis result.

As described above, the document generation device 10 functions as an example of an output determination unit that determines whether or not to output the message based on the analysis result.

When generating a message (step S5; YES), the document generator 10 generates a message (step S6). Specifically, the system control unit 16 (message generation unit 23) sets the e-mail address to "tanaka@...c.ne.jp" as shown in FIG. The content of the email (an example of a message) is generated by adding the document "Display map" that links to the map information indicating the current location as additional information to "(an example of a document)." As additional information, only the map may be displayed under the document "I am passing here now."

As described above, the document generation device 10 functions as an example of a message generation unit that generates a message according to the document and the additional information.

Next, the document generator 10 outputs a message (step S7). Specifically, the output unit 15 transmits the generated message via the communication unit 13 based on the mail address. The output unit 15 may display the generated message on the display of the document generation device 10.

As described above, the document generation device 10 functions as an example of the output unit that outputs the message to the output destination based on the determination result of the output determination unit.

If the message is not generated (step S5; NO), the document generation device 10 ends the process.

(2.2 Subroutine for analysis of morphological analysis results)
Next, a subroutine for analyzing the morphological analysis result will be described with reference to FIG.

As shown in FIG. 10, the document generator 10 acquires the morphological analysis result (step S10). Specifically, the system control unit 16 (document analysis unit 22) acquires the processing result of step S3.

Next, the document generation device 10 determines whether or not a keyword meaning transmission exists (step S11). Specifically, the system control unit 16 (output keyword extraction unit 22b) refers to the output database of the storage unit 12 and transmits "contact", "send", "mail", etc. in the decomposed morphemes. If the word that means is extracted and can be extracted, it is determined that the keyword that means transmission exists. In the case of the example of FIG. 9, since there is a morpheme of "contact", the system control unit 16 determines that a keyword meaning transmission exists.

As described above, the document generation device 10 functions as an example of an extraction unit that extracts keywords from the text data according to the analysis result. Further, the document generation device 10 functions as an example of an output determination unit that determines whether or not to output the message based on the analysis result.

When the keyword meaning transmission exists (step S11; YES), the document generator 10 determines whether or not the name information of the output destination exists (step S12). Specifically, the system control unit 16 (output destination extraction unit 22c) refers to the address database of the storage unit 12 (an example of the output destination database), and puts the decomposed morpheme into the “name” of the address database. Determine if the corresponding information exists. In the case of the example of FIG. 9, since there is a morpheme "Tanaka", the system control unit 16 determines that the name information of the output destination exists.

As described above, the document generation device 10 functions as an example of the output destination extraction unit that extracts the output destination from the output destination database in which the output destination of the message is registered according to the analysis result.

As described above, the document generation device 10 functions as an example of an extraction unit that extracts keywords from the text data according to the analysis result. Further, the document generation device 10 functions as an example of an output determination unit that determines whether or not to output the message based on the analysis result.

When the name information of the output destination exists (step S12; YES), the document generator 10 determines whether or not a matching keyword exists (step S13). Specifically, the system control unit 16 (keyword extraction unit 22e) refers to the keyword database of the storage unit 12 and determines whether or not there is a morpheme that matches the keyword of the keyword database among the decomposed morphemes. do. In the case of the example of FIG. 9, since there is a morpheme of "current location", the system control unit 16 determines that a matching keyword exists.

As described above, the document generation device 10 functions as an example of an extraction unit that extracts keywords from the text data according to the analysis result. Further, the document generation device 10 functions as an example of an output determination unit that determines whether or not to output the message based on the analysis result.

Document generation when there is no keyword meaning transmission (step S11; NO), when the output destination name information does not exist (step S12; NO), or when there is no matching keyword (step S13; NO). The device 10 sets the message ungenerated flag (step S14). Specifically, the system control unit 16 sets a flag area for message generation in the RAM 16c, and sets the flag to "0".

When a matching keyword exists (step S13; YES), the document generation device 10 acquires the address information and the fixed phrase information (step S15). Specifically, the system control unit 16 (output destination extraction unit 22c) refers to the address database of the storage unit 12 and acquires an e-mail address as address information. In the case of the example of FIG. 9, the system control unit 16 acquires the e-mail address information "tanaka@...c.ne.jp" from the address database based on the name information "Tanaka". Further, the system control unit 16 (keyword extraction unit 22e) refers to the keyword database of the storage unit 12 and acquires fixed phrase information. In the case of the example of FIG. 9, the system control unit 16 acquires the fixed phrase "passing here and now" from the keyword base based on the keyword "current location" and generates a document.

As described above, the document generation device 10 functions as an example of the document generation unit that generates a document according to the result of analyzing the text data converted from the voice signal spoken by the speaker.

Next, the document generation device 10 determines whether or not a keyword or the like for additional information exists (step S16). Specifically, the system control unit 16 (additional information generation unit 22g) refers to the additional information database of the storage unit 12, and refers to the keyword for additional information or the fixed form corresponding to the fixed phrase acquired in step S15. Determines if the statement number exists.

As described above, the document generation device 10 functions as an example of an extraction unit that extracts keywords from the text data according to the analysis result.

When the keyword or the like for the additional information exists (step S16; YES), the document generation device 10 acquires the additional information and the fixed phrase information for the additional information (step S17). The system control unit 16 (additional information generation unit 22g) refers to the additional information database of the storage unit 12 and acquires the additional information corresponding to the keyword for the additional information and the additional information corresponding to the fixed phrase number. In the case of the example of FIG. 9, the system control unit 16 acquires the item of the map information as the additional information from the additional information database based on the number of the fixed phrase corresponding to the “current location”. Further, the system control unit 16 acquires fixed phrase information (“map display”) corresponding to the additional information “map information”. The system control unit 16 sets a flag indicating that there is additional information.

As described above, the document generation device 10 functions as an example of the additional information generation unit that generates additional information to be added to the document based on at least one of the document and the keyword. Further, the document generation device 10 is an additional information generation unit that acquires information related to the current location and generates the additional information including information related to the current location based on at least one of the document and the keyword. Works as an example.

Next, the document generation device 10 sets the message generation flag (step S18). Specifically, the system control unit 16 sets a flag area for message generation in the RAM 16c, and sets the flag to "1".

If the keyword or the like for additional information does not exist (step S16; NO), the message generation flag is set (step S18). Then, the system control unit 16 sets a flag indicating that there is no additional information.

When the subroutine ends, the document generation device 10 holds the value of the flag and performs the process of step S5.

(2.3 Message generation subroutine)
Next, a subroutine for generating a message will be described with reference to FIG.

As shown in FIG. 11, the document generator 10 sets the acquired address information (step S20). Specifically, the system control unit 16 sets the e-mail address of the e-mail software.

Next, the document generation device 10 acquires phrases other than those related to keywords and additional information (step S21). For example, in the case of the example of FIG. 9, the system control unit 16 (message generation unit 23) acquires morphemes other than the keyword “Tanaka”, the output keyword “contact”, and the keyword “current location”. Then, the system control unit 16 extracts morphemes that can be used for the message. In the case of the example of FIG. 9, if these keywords are excluded, there may be no words that can be used in the message.

Next, the document generation device 10 generates a basic message from clauses and fixed phrases other than those related to keywords and additional information (step S22). For example, in the case of the example of FIG. 9, the system control unit 16 (message generation unit 23). ) Generates the basic message "I'm passing here now".

Next, the document generation device 10 determines whether or not to acquire additional information or the like (step S23). Specifically, the system control unit 16 (message generation unit 23) determines in step S16 that additional information or the like is to be acquired when a flag is set to indicate that there is additional information.

When it is determined to acquire the additional information or the like (step S23; YES), the document generation device 10 adds the additional information or the like to the basic message (step S24). Specifically, the system control unit 16 (message generation unit 23) adds a document called "map display" related to additional information and map information (latitude, longitude) to the basic message to generate a message. ..

If it is not determined to acquire the additional information or the like (step S23; NO), the document generation device 10 proceeds to step S7 without adding the additional information. For example, as shown in FIG. 12, in the case of the voice recognition result "Email Tanaka 10 minutes late", there is no keyword corresponding to the additional information. Note that "mail" is an output keyword, "delayed" is a keyword related to a fixed phrase, "Tanaka" is name information, and "10" and "minute" are examples of phrases other than those related to additional information.

Next, other speech recognition results will be described with reference to FIGS. 13 and 14.

As shown in FIG. 13, "mail" is a keyword meaning transmission, "Tanaka" is a keyword for name information of an output destination, and "delayed" is a keyword for extracting a fixed phrase, "traffic jam". Is a keyword for additional information.

In step S17, the document generator 10 acquires the current position from the sensor unit 14 from the keyword "traffic jam" for additional information, and acquires the traffic jam information via the communication unit 13 based on the route information to the destination. do. Then, the document generation device 10 generates a fixed phrase "Kannana-dori XX to 2 km traffic jam" for additional information from the traffic jam information and the current location information. Here, "○○" is the current position.

The document generation device 10 generates a message having mail contents as shown in FIG.

As described above, the document generation device 10 functions as an example of the additional information generation unit that generates the additional information including the information about the current location to the destination.

As shown in FIG. 14, "send" is a keyword meaning transmission, "Tanaka" is a keyword for name information of the output destination, and "wait" is a keyword for extracting a fixed phrase. "Station" and "wait" are keywords for additional information.

In step S17, the document generation device 10 acquires map information near the station from the keyword for additional information.

The document generation device 10 generates a message having mail contents as shown in FIG.

As described above, according to the operation according to the embodiment, the words spoken by the speaker can be complemented with additional information to generate a natural message. It also facilitates the creation of documents for the speaker.

In addition, when the current location is acquired, the information related to the current location is acquired, and the additional information including the information related to the current location is generated based on at least one of the document and the keyword, the information about the current location is added. , It is possible to convey accurate information by complementing the information to be conveyed to the other party.

When generating additional information including information about the current location to the destination, information regarding the current location to the destination (on the route), for example, traffic jam information, arrival time, and weather information can be requested and added to the message. It makes it easier for the speaker to create a document without having to bother to search for traffic jam information or weather information.

In addition, when extracting the output destination from the output destination database (for example, the address database) in which the output destination of the message is registered according to the analysis result, it is not necessary to input the output destination, and the speaker's operation can be performed. It will be easier.

When it is determined whether or not to output a message based on the analysis result and the message is output to the output destination based on the determination result, it is not necessary to input the output destination and the operation of the speaker becomes easy.

10: Document generation device 10a: Document generation unit 10b: Extraction unit 10c: Additional information generation unit 10d: Message generation unit 11: Input unit 14: Sensor unit 15: Output unit 16: System control unit (Document generation unit, extraction unit, Additional information generation unit, message generation unit)
21: Voice recognition unit
22: Document analysis unit 22a: Morphological analysis unit 22b: Output keyword extraction unit 22c: Output destination extraction unit 22e: Keyword extraction unit 22f: Document generation unit 22g: Additional information generation unit 23: Message generation unit

Claims (1)

The first acquisition unit that acquires the text data in which the voice spoken by the speaker is converted by voice recognition, and
An extraction unit that extracts keywords from the text data,
The second acquisition unit that acquires the fixed phrase information corresponding to the keyword, and
An additional information generation unit that generates additional information corresponding to at least one of the fixed phrase information and the keyword from the additional information database that stores the additional information.
A message generation unit that generates a message based on the fixed phrase information and the additional information,
A document generator characterized by being equipped with.

Images