JP7348818B2 - dialogue device - Google Patents

Description

TECHNICAL FIELD The present invention relates to a dialogue device.

2. Description of the Related Art Dialogue devices that respond to text input or utterances from a user are known (for example, see Patent Document 1). Such a dialog device can provide answers to questions from users according to scenarios prepared in advance.

JP 2019-133418 Publication

Here, with the above-mentioned dialogue device, it is difficult to appropriately answer questions for which no scenario has been prepared. This may make the user feel that there are few variations in dialogue and that it is not easy to use, and there is a risk that the use of the dialogue device cannot be sufficiently promoted.

The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to provide an interaction device that can promote user usage.

An interaction device according to one aspect of the present invention includes: a wrong answer determining unit that determines an incorrect answer term that is similar in pronunciation to an input term that is a term input by a user or a term related to the term; The present invention includes a correct answer determining unit that determines a correct answer term that is similar to the correct answer term, a question sentence generating unit that generates a question sentence using the correct answer term as an answer, and an output unit that outputs the question sentence.

In the dialog device according to one aspect of the present invention, when a user inputs a term, a question based on the input term is generated and output. As a result, even if it is not possible to appropriately answer a question from a user, for example, any question related to the content of the user's question (terminology input by the user) will not be output from the dialog device. , the interactive dialogue between the user and the dialogue device continues. Thereby, it is possible to improve the user's sense of familiarity with the dialogue device and to improve the usability of the dialogue device. In addition, specifically, the dialogue device determines an incorrect answer term that is similar in pronunciation to the input term, determines a correct answer term that is similar in meaning to the incorrect answer term, and generates a question sentence with the correct answer term as an answer. Output. Since the incorrect answer term is similar in pronunciation to the input term and similar in meaning to the correct answer term of the question, the user's answer to the question text can be effectively guided to the incorrect answer term. By doing this, it is possible to improve the gameplay in a series of dialogues such as terminology input from the user, questions to the user, and answers from the user, thereby improving the user's sense of familiarity with the dialogue device and promoting its use. Can be done.

According to the present invention, it is possible to promote the user's use of the dialogue device.

FIG. 2 is a block diagram showing the functional configuration of a dialogue device included in the dialogue system according to the present embodiment. It is a figure explaining an input term, an incorrect answer term, and a correct answer term. It is a flowchart which shows the process which an interaction device performs. FIG. 2 is a diagram showing the hardware configuration of an interaction device.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same reference numerals are used for the same or equivalent elements, and overlapping description will be omitted.

FIG. 1 is a block diagram showing the functional configuration of a dialogue device 10 included in a dialogue system 1 according to the present embodiment. The dialogue system 1 shown in FIG. 1 is a system that provides answers to inquiries from a user terminal 50 (user). The user terminal 50 may be any communication terminal, such as a smartphone, a tablet terminal, a PC, or the like. In the following description, it is assumed that the user terminal 50 is a terminal capable of making phone calls. In the dialog system 1, a dialog between the user terminal 50 and the dialog device 10 is started when the dialog device 10 receives an inquiry from the user terminal 50. In the dialog system 1, for example, if the dialog device 10 is capable of responding to an inquiry from the user terminal 50, the response is provided from the dialog device 10 to the user terminal 50, and the dialog responds to the inquiry from the user terminal 50. If the device 10 is not able to answer, the question is switched to a predetermined game mode and the question is output from the dialog device 10 to the user terminal. The question here is not a question for providing an answer to an inquiry from the user terminal 50, but a question related to an interactive game (details will be described later).

The dialogue device 10 is a device that provides answers to inquiries from the user terminal 50 by having a dialogue with the user terminal 50 . Furthermore, the interaction device 10 is a device that provides an interactive game to the user terminal 50. That is, the dialog device 10 is configured to be able to switch between a normal mode in which answers are provided to the user terminal 50 and a game mode in which a dialog game is provided to the user terminal 50. The interactive game is a game provided to the user terminal 50 through dialogue between the dialogue device 10 and the user terminal 50, and includes, for example, I. II. The dialog device 10 prompts the user terminal 50 to repeatedly utter a certain term (for example, "chandelier"). The user repeatedly utters a certain term, III. IV. The dialogue device 10 asks a question to the user terminal 50 (for example, "Who died after eating the poisoned apple given by the Queen?"). This is a game that is played as the user answers questions. In the above example, the user who repeatedly said ``chandelier'' would respond to the question ``Who died after eating the poisoned apple the queen gave him?'' when he should have answered ``Snow White'' instead. It is conceivable that due to the influence of repeated utterances, etc., the answer is "Cinderella." In interactive games, the gameplay is improved by inducing such incorrect answers. The dialogue device 10 is a device using a so-called chatbot. Chatbot is a term that combines chat and bot, and is an automatic dialogue program that provides information to the user while having a dialogue with the user, using, for example, artificial intelligence.

As shown in FIG. 1, the dialogue device 10 includes an input section 11, an answer determination section 12, an information DB 13, a score derivation section 14, an incorrect answer determination section 15, a correct answer determination section 16, and a question sentence generation section. It includes a section 17 and an output section 18.

The input unit 11 receives an incoming call from the user terminal 50 and acquires an inquiry from the user terminal 50 . The inquiry is obtained, for example, according to the result of voice recognition regarding the utterance of the user terminal 50. Note that voice recognition is performed using a conventionally known technique. Speech recognition may be performed in the dialogue device 10, or may be performed by an external device (not shown) and the dialogue device 10 acquires the result. The input unit 11 outputs the inquiry acquired from the user terminal 50 to the answer determining unit 12.

The answer determining unit 12 receives input of an inquiry from the input unit 11 and determines an answer to the inquiry. The answer determining unit 12 refers to the scenario stored in the information DB 13 and determines whether it is possible to answer the inquiry based on the scenario. If an answer to the inquiry is defined in the scenario, the answer determining unit 12 determines the answer to the inquiry based on the scenario. In this case, the answer determining section 12 outputs the answer to the output section 18. On the other hand, if the answer to the inquiry is not specified in the scenario and cannot be answered, the answer determining unit 12 outputs the inquiry of the user terminal 50 to the incorrect answer determining unit 15.

The information DB 13 is a database that stores various information used in processing of the dialog device 10. The information DB 13 stores scenarios that define answers to inquiries. The information DB 13 also stores a word dictionary that defines information regarding the pronunciations and meanings of a plurality of words. Note that the interaction device 10 does not need to have the information DB 13. In this case, the dialogue device 10 acquires various information stored in the information DB 13 described above from an external database.

The incorrect answer determining unit 15 receives an inquiry input from the answer determining unit 12, identifies an input term that is a term input from the user terminal 50 or a term related to the term based on the inquiry, and assigns the input term to the input term. Determine incorrect words with similar pronunciations. For example, as shown in FIG. 2, when the input term is "chandelier," an example of the incorrect word is "Cinderella." The term input from the user terminal 50 is the term itself included in the inquiry. Terms related to the terms input from the user terminal 50 include, for example, terms that are a combination of the terms included in the inquiry and another term, terms that are more or less specific than the terms included in the inquiry, and terms that are included in the inquiry. These are terms that are closely related to each other.

The incorrect answer determining unit 15 captures, for example, a sentence related to an inquiry using a regular expression. Alternatively, the incorrect answer determination unit 15 divides the sentence related to the inquiry into morphemes and characters using a model constructed based on machine learning (for example, a method such as CRF (conditional random field)), and then processes the input term identification process. I do. Alternatively, the incorrect answer determining unit 15 may specify the input term by extracting one term from the sentence that completely matches a word in the incorrect answer candidate dictionary. Thereafter, when estimating the reading of the input term, if there is a reading in the incorrect answer candidate dictionary, the incorrect answer determining unit 15 may use the reading as the reading of the input term. Further, when there is no reading in the incorrect answer candidate dictionary, the incorrect answer determining unit 15 may estimate the reading of the input term. The incorrect answer determining unit 15 decomposes the input term into subwords, for example by performing morphological analysis, and estimates the pronunciation of the input term based on the pronunciation of the subwords. The incorrect answer determination unit 15 converts the input term into Hepburn Roman letters, for example, and refers to the word dictionary stored in the information DB 13 to extract a plurality of incorrect answer term candidates that are similar in pronunciation to the input term. Note that the incorrect answer determining unit 15 may estimate a term that is not in the word dictionary as an incorrect answer term candidate. The incorrect answer determining unit 15 outputs the extracted or estimated incorrect answer term candidates to the score deriving unit 14. The incorrect answer determining unit 15 then obtains the score of each incorrect answer term candidate (score regarding validity as an incorrect answer term) from the score deriving unit 14, and selects a plurality of incorrect answers based on the score derived by the score deriving unit 14. Determine the incorrect term from the term candidates. The incorrect answer determining unit 15 determines, for example, the incorrect answer term candidate with the lowest score as the incorrect answer term. The incorrect answer determining unit 15 outputs the determined incorrect answer term to the correct answer determining unit 16. Further, the incorrect answer determining unit 15 outputs the specified input term to the score deriving unit 14 and the output unit 18.

The correct answer determining unit 16 receives input of the incorrect answer term from the incorrect answer determining unit 15, and determines a correct answer term that is similar in meaning to the incorrect answer term. For example, as shown in FIG. 2, when the incorrect answer term is "Cinderella", an example of the correct answer term is "Snow White". The correct answer determining unit 16 refers to the word dictionary stored in the information DB 13 and extracts a plurality of correct answer term candidates that are considered to be similar in meaning to the incorrect answer term. Note that the correct answer determining unit 16 may estimate a term that is not in the word dictionary as a correct answer term candidate. The correct answer determining unit 16 outputs the extracted or estimated correct answer term candidates to the score deriving unit 14. Then, the correct answer determining unit 16 obtains the score of each correct answer term candidate (score regarding validity as a correct answer term) from the score deriving unit 14, and selects the correct answer from the plurality of correct answer term candidates based on the score derived by the score deriving unit 14. Decide on terminology. The correct answer determining unit 16 determines, for example, the correct answer term candidate with the lowest score as the correct answer term. The correct answer determining unit 16 outputs the determined correct answer term to the question sentence generating unit 17.

The score derivation unit 14 derives scores regarding the validity of the plurality of incorrect answer term candidates input from the incorrect answer determination unit 15 as incorrect answer terms. The score deriving unit 14 outputs the derived score of each incorrect answer term candidate to the incorrect answer determining unit 15. The score derivation unit 14 derives a score regarding the validity of each of the plurality of incorrect word candidates as an incorrect word by calculating, for example, the phonetic distance from the input term. Phonetic distance indicates the closeness (similarity) of pronunciation. For example, the score derivation unit 14 calculates, for each of the plurality of incorrect word candidates, the phonetic distance between the input term and consonants excluding n, and the phonetic distance between the vowel, n, and long consonant with the input term. Obtained using Stein distance. The score derivation unit 14 derives the score of the incorrect word candidate by using, for example, the sum of the two types of phonetic distances as the phonetic distance between the input term and the incorrect word candidate. Therefore, the closer the acoustic distance, the lower the score. Note that the score deriving unit 14 may derive the score of the incorrect term candidate by deriving the phonetic distance between the input term and the incorrect term candidate, without separating consonants and vowels.

For example, as shown in FIG. 2, a case will be explained in which the input term is "chandelier" and the incorrect term candidate is "Cinderella." In this case, the acoustic distance between consonants excluding n is derived by comparing "shdr" (chandelier) and "shdrr" (Cinderella). Further, the phonetic distance between vowels, n, and long sounds is derived by comparing "aneia" (chandelier) and "ineea" (Cinderella).

The score deriving unit 14 calculates the phonetic distance between consonants except for n, applying a bias based on, for example, the IPA consonant table. Specifically, for example, the acoustic distance between p and b is short, and the acoustic distance between p and h is long. In addition, the score deriving unit 14 calculates the phonetic distance between vowels, n, and long sounds using a bias based on, for example, a basic vowel table. Specifically, for example, the acoustic distance between ie is short and the acoustic distance between au is long. For long sounds, the score deriving unit 14 uses, for example, the phonetic distance of the previous vowel. Note that the score derivation unit 14 may derive the score of the incorrectly answered term by comparing the audio waveforms of the input term and the incorrectly answered term candidate.

The score deriving unit 14 derives a score regarding the validity as a correct term for the plurality of correct term candidates input from the correct answer determining unit 16. The score derivation unit 14 outputs the derived score of each correct answer term candidate to the correct answer determination unit 16. The score derivation unit 14 derives a score regarding the validity of each of the plurality of correct term candidates as a correct term, by calculating the semantic distance from the incorrect term inputted from the incorrect answer determining unit 15, for example. Semantic distance indicates the closeness (similarity) of meaning. The score deriving unit 14 derives the score of each correct term candidate by deriving the semantic distance of each correct term candidate from the incorrect term based on, for example, the cosine similarity based on the word vector or the tf-idf value. Therefore, the closer the semantic distance, the lower the score.

The score deriving unit 14 calculates the degree of popularity (degree of widespread use) of a term, and calculates a score regarding the validity of each of the plurality of incorrect term candidates as an incorrect term, and a score for each of the plurality of correct term candidates. A score regarding the validity of the term as a correct answer term may be derived. The popularity of a term can be calculated, for example, based on the frequency of appearance in a corpus (a collection of sentences). For example, the score derivation unit 14 lowers the score for a more popular incorrect term candidate (correct term candidate) to make it more likely to be determined as an incorrect term (correct term). The score deriving unit 14 derives scores of a plurality of incorrect answer term candidates and scores of a plurality of correct answer term candidates, for example, based on the following equation (1). α and β are adjustable parameters.
Score = α・log (phonological distance) + β・log (semantic distance) − γlog (term popularity) (1)

The score derivation unit 14 makes α in the above formula (1) sufficiently larger than β when deriving the score of the incorrect answer term candidate, and when deriving the score of the correct answer term candidate, the score derivation unit 14 sets the score ( 1) Hyperparameters in equation (1) are appropriately adjusted, such as by making β in equation sufficiently larger than α, and scores for incorrect term candidates and correct term candidates are derived. Note that the sum or product of the score of the correct answer term determined by the correct answer determining unit 16 and the score of the incorrect answer term determined by the incorrect answer determining unit 15 is the sum or product of the score of the correct answer term determined by the correct answer determining unit 16. It may be used as a game quality score. The game quality score is derived, for example, based on the following equation (2). X and Y are adjustable parameters.
Game quality score = X, correct answer term score + Y, incorrect answer term score... (2)

For example, it is determined that the lower the game quality score, the higher the game quality. That is, the closer the phonetic distance between the incorrect answer term and the input term, the closer the semantic distance between the correct answer term and the incorrect answer term, and the higher the popularity of the incorrect answer term and the correct answer term, the higher the game quality.

The question generation unit 17 generates a question whose answer is the correct answer term input from the correct answer determination unit 16. For example, the question generation unit 17 uses a search function to extract sentences related to the correct answer term from the Internet, and generates a question based on the sentences. The question sentence generation unit 17 extracts sentences related to correct answer terms from, for example, encyclopedias, dictionaries, example collections, etc. on the Internet. Assume now that the correct term is "Snow White" as shown in FIG. In this case, for example, suppose that the following sentence is extracted from the example collection: ``...But it was too late, and Snow White ate the poisoned apple that the queen gave her and died.'' The question sentence generation unit 17 deletes the subject from such a sentence or adds ``no wa?'' to the end of the sentence to make it a question sentence, so that, for example, ``I took a breath after eating the poisoned apple that the queen gave me.'' Generates a question such as "What has been extinct?" The question generation unit 17 outputs the generated question to the output unit 18.

The output unit 18 outputs information in response to user utterances. Specifically, when the answer to the user's inquiry has been determined by the answer determining unit 12, the output unit 18 outputs the answer to the user terminal 50 by voice or the like. Furthermore, when the question text generation unit 17 generates a question text in the game mode, the output unit 18 outputs the generated question text to the user terminal 50 by voice or the like. Furthermore, in the game mode, the output unit 18 first outputs to the user terminal 50 a repeated utterance guidance sentence for repeatedly (for example, 10 times) uttering the input term specified by the incorrect answer determining unit 15, For example, the above-mentioned question sentence may be output after the input term is repeatedly uttered by the user.

The flow of the game mode in this case is as follows. First, the output unit 18 outputs a repeated utterance guidance sentence to the user terminal 50, such as ``Try saying chandelier ten times'' in order to have the user repeatedly utter the input term (``chandelier''). In response, the user of the user terminal 50 utters "chandelier" ten times. After that, the output unit 18 outputs the question "Who died after eating the poisoned apple given by the queen?". The correct answer (correct answer term) is "Snow White," but a user who has repeatedly said "chandelier" in the previous sentence is likely to answer the incorrect answer (incorrect answer term) as "Cinderella." In this way, by repeating the input terms in the previous stage, the gameplay can be improved.

Next, the flow of processing performed by the dialog device 10 will be explained with reference to the flowchart of FIG. 3. As shown in FIG. 3, the dialogue device 10 first receives an utterance from the user terminal 50 (step S1). The dialogue device 10 obtains a user's inquiry according to the result of voice recognition regarding the utterance, for example.

Next, the dialog device 10 determines whether an answer to the inquiry is possible (step S2). Specifically, the dialog device 10 refers to a scenario stored in the information DB 13 and determines whether or not it is possible to answer the inquiry based on the scenario.

In step S2, if an answer to the inquiry is possible (there is an answer in the scenario), the dialogue device 10 specifies the answer to the inquiry based on the scenario and outputs the answer to the user terminal 50 (step S3). .

On the other hand, in step S2, if an answer to the inquiry is not possible (there is no answer in the scenario), the interaction device 10 provides the user terminal 50 with an interactive game from the normal mode in which answers are provided to the user terminal 50. Switch to the game mode you want to use and perform the game mode processing. Specifically, the dialog device 10 first identifies input terms from the user terminal 50 based on the inquiry, and determines incorrectly answered terms from the input terms (step S4). The dialogue device 10 extracts a plurality of incorrect answer term candidates that are similar in pronunciation to the input term, derives a score regarding the validity of each incorrect answer term candidate as an incorrect answer term, and selects an incorrect answer term based on the score. Determine.

Continuing, the dialog device 10 determines the correct answer term from the incorrect answer terms (step S5). The dialogue device 10 extracts a plurality of correct term candidates that are similar in meaning to the incorrect term, derives a score regarding the validity of each correct term candidate as a correct term, and determines the correct term based on the score. .

Subsequently, the dialogue device 10 generates a question sentence whose answer is the correct answer term (step S6). The dialog device 10 extracts a document related to the correct answer term from the Internet using, for example, a search function, and generates a question based on the document. Then, the dialogue device 10 outputs the generated question text to the user terminal 50 by voice or the like. Note that in the game mode, the dialogue device 10 first outputs to the user terminal 50 a repeated utterance guidance sentence for causing the input term to be uttered repeatedly (for example, 10 times), and for example, when the input term is repeatedly uttered by the user. After that, the above-mentioned question text may be output.

Next, the effects of the dialogue device 10 according to this embodiment will be explained.

The dialogue device 10 according to the present embodiment includes a wrong answer determining unit 15 that determines an incorrect answer term that is similar in pronunciation to an input term that is a term input from the user terminal 50 or a term related to the term, and an incorrect answer term The present invention includes a correct answer determining unit 16 that determines a correct answer term similar in meaning to , a question sentence generating unit 17 that generates a question sentence using the correct answer term as an answer, and an output unit 18 that outputs the question sentence.

In such an interaction device 10, when a certain term is input from the user terminal 50, a question based on the input term is generated and output. For example, even if a question from the user terminal 50 cannot be answered appropriately, any question related to the content of the question from the user terminal 50 (terms input from the user terminal 50) is output from the dialogue device 10, and the interactive dialogue between the user terminal 50 and the dialogue device 10 continues. Thereby, the user's sense of familiarity with the dialogue device 10 can be improved, and the usability of the dialogue device 10 can be improved. In addition, in the dialogue device 10, specifically, an incorrect answer term that is similar in pronunciation to the input term is determined, a correct answer term that is similar in meaning to the incorrect answer term is determined, and a question sentence whose answer is the correct answer term is generated. and output. Since the incorrect answer term is similar in pronunciation to the input term and similar in meaning to the correct answer term of the question, the user's answer to the question text can be effectively guided to the incorrect answer term. This makes it possible to improve the gameplay in a series of dialogues including terminology input from the user, questions to the user, and answers from the user, thereby improving the user's sense of familiarity with the dialogue device 10 and promoting its use. be able to.

The dialogue device 10 further includes a score derivation unit 14 that derives a score regarding the validity of a plurality of incorrect answer term candidates as an incorrect answer term, and derives a score regarding the validity of a plurality of correct answer term candidates as a correct answer term. , the incorrect answer determining unit 15 determines an incorrect answer term from a plurality of incorrect answer term candidates based on the score derived by the score deriving unit 14, and the correct answer determining unit 16 determines an incorrect answer term from a plurality of incorrect answer term candidates based on the score derived by the score deriving unit 14. Determine the correct answer term from the correct answer term candidates. In this way, by determining the incorrect answer term and the correct answer term based on the scores of the plurality of candidates, it is possible to select the appropriate incorrect answer term and correct answer term based on a certain criterion.

The score derivation unit 14 derives a score regarding the validity of each of the plurality of incorrect term candidates as an incorrect term by calculating the phonetic distance from the input term. By considering whether the sounds are similar (the phonetic distance is small), incorrect words with similar pronunciations can be appropriately derived.

The score derivation unit 14 derives a score regarding the validity of each of the plurality of correct term candidates as a correct answer candidate by calculating the semantic distance from the incorrect term. By considering whether the semantic distance is small, it is possible to appropriately derive a correct answer term that is similar in meaning to an incorrect answer term.

By calculating the popularity of the term, the score derivation unit 14 calculates a score regarding the validity of each of the plurality of incorrect answer term candidates as an incorrect answer term, and a score regarding the validity of each of the plurality of correct answer term candidates as a correct answer candidate. Derive the score. Thereby, for example, a popular term can be easily selected as an incorrect answer term or a correct answer term, and user satisfaction can be improved. In addition, by using a popular term as an incorrect answer term, the user's answer to the question can be effectively guided to the incorrect answer term, and the gameplay can be subtracted.

The question sentence generation unit 17 uses a search function to extract sentences related to the correct answer term, and generates a question sentence based on the sentences. Thereby, it is possible to easily and appropriately generate a question text whose answer is the correct answer term.

The output unit 18 outputs a question sentence after outputting a repeated utterance guidance sentence for causing the input term to be repeatedly uttered. Thereby, when a question sentence is output, it is possible to make it easier to answer with incorrectly answered terms.

Finally, the hardware configuration of the dialogue device 10 included in the dialogue system 1 will be explained with reference to FIG. 4. The above-described interaction device 10 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

In addition, in the following description, the word "apparatus" can be read as a circuit, a device, a unit, etc. The hardware configuration of the dialog device 10 may be configured to include one or more of each device shown in the figure, or may be configured not to include some of the devices.

Each function in the dialogue device 10 is performed by loading predetermined software (programs) onto hardware such as a processor 1001 and a memory 1002, so that the processor 1001 performs calculations, performs communication by the communication device 1004, and performs communication using the memory 1002 and storage 1003. This is realized by controlling reading and/or writing of data in.

The processor 1001, for example, operates an operating system to control the entire computer. The processor 1001 may be configured with a central processing unit (CPU) that includes interfaces with peripheral devices, a control device, an arithmetic device, registers, and the like. For example, the control functions of the incorrect answer determining unit 15 and the like of the dialogue device 10 may be realized by the processor 1001.

Further, the processor 1001 reads programs (program codes), software modules, and data from the storage 1003 and/or the communication device 1004 to the memory 1002, and executes various processes in accordance with these. As the program, a program that causes a computer to execute at least part of the operations described in the above embodiments is used. For example, the control functions of the wrong answer determining unit 15 and the like of the dialogue device 10 may be realized by a control program stored in the memory 1002 and operated by the processor 1001, and may be similarly realized with respect to other functional blocks. . Although the various processes described above have been described as being executed by one processor 1001, they may be executed by two or more processors 1001 simultaneously or sequentially. Processor 1001 may be implemented with one or more chips. Note that the program may be transmitted from a network via a telecommunications line.

The memory 1002 is a computer-readable recording medium, and includes at least one of ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. may be done. Memory 1002 may be called a register, cache, main memory, or the like. The memory 1002 can store executable programs (program codes), software modules, and the like to implement a wireless communication method according to an embodiment of the present invention.

The storage 1003 is a computer-readable recording medium, such as an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (such as a compact disk, a digital versatile disk, or a Blu-ray disk). (registered trademark) disk), smart card, flash memory (eg, card, stick, key drive), floppy disk, magnetic strip, etc. Storage 1003 may also be called an auxiliary storage device. The storage medium mentioned above may be, for example, a database including memory 1002 and/or storage 1003, a server, or other suitable medium.

The communication device 1004 is hardware (transmission/reception device) for communicating between computers via a wired and/or wireless network, and is also referred to as, for example, a network device, network controller, network card, communication module, or the like.

The input device 1005 is an input device (eg, keyboard, mouse, microphone, switch, button, sensor, etc.) that accepts input from the outside. The output device 1006 is an output device (for example, a display, a speaker, an LED lamp, etc.) that performs output to the outside. Note that the input device 1005 and the output device 1006 may have an integrated configuration (for example, a touch panel).

Further, each device such as the processor 1001 and the memory 1002 is connected by a bus 1007 for communicating information. The bus 1007 may be configured as a single bus or may be configured as different buses between devices.

The dialogue device 10 also includes hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), and a field programmable gate array (FPGA). A part or all of each functional block may be realized by the hardware. For example, processor 1001 may be implemented with at least one of these hardware.

Although this embodiment has been described in detail above, it is clear for those skilled in the art that this embodiment is not limited to the embodiment described in this specification. This embodiment can be implemented as modifications and changes without departing from the spirit and scope of the present invention as defined by the claims. Therefore, the description in this specification is for the purpose of illustrative explanation and does not have any restrictive meaning with respect to this embodiment.

Each aspect/embodiment described herein applies to LTE (Long Term Evolution), LTE-A (LTE-Advanced), SUPER 3G, IMT-Advanced, 4G, 5G, FRA (Future Radio Access), W-CDMA. (registered trademark), GSM (registered trademark), CDMA2000, UMB (Ultra Mobile Broad-band), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, UWB (Ultra-Wide) The present invention may be applied to systems utilizing Bluetooth (registered trademark), Bluetooth (registered trademark), and other appropriate systems, and/or next-generation systems expanded based on these.

The order of the processing procedures, sequences, flowcharts, etc. of each aspect/embodiment described in this specification may be changed as long as there is no contradiction. For example, the methods described herein present elements of the various steps in an exemplary order and are not limited to the particular order presented.

The input/output information may be stored in a specific location (eg, memory) or may be managed in a management table. Information etc. to be input/output may be overwritten, updated, or additionally written. The output information etc. may be deleted. The input information etc. may be transmitted to other devices.

Judgment may be made using a value expressed by 1 bit (0 or 1), a truth value (Boolean: true or false), or a comparison of numerical values (for example, a predetermined value). (comparison with a value).

Each aspect/embodiment described in this specification may be used alone, may be used in combination, or may be switched and used in accordance with execution. In addition, notification of prescribed information (for example, notification of "X") is not limited to being done explicitly, but may also be done implicitly (for example, not notifying the prescribed information). Good too.

Software includes instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, whether referred to as software, firmware, middleware, microcode, hardware description language, or by any other name. , should be broadly construed to mean an application, software application, software package, routine, subroutine, object, executable, thread of execution, procedure, function, etc.

Additionally, software, instructions, etc. may be sent and received via a transmission medium. For example, if the software uses wired technologies such as coaxial cable, fiber optic cable, twisted pair and digital subscriber line (DSL) and/or wireless technologies such as infrared, radio and microwave to When transmitted from a remote source, these wired and/or wireless technologies are included within the definition of transmission medium.

The information, signals, etc. described herein may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc., which may be referred to throughout the above description, may refer to voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, light fields or photons, or any of these. It may also be represented by a combination of

Note that terms explained in this specification and/or terms necessary for understanding this specification may be replaced with terms having the same or similar meanings.

Further, the information, parameters, etc. described in this specification may be expressed as absolute values, relative values from a predetermined value, or other corresponding information. .

A user terminal is defined by a person skilled in the art as a mobile communication terminal, a subscriber station, a mobile unit, a subscriber unit, a wireless unit, a remote unit, a mobile device, a wireless device, a wireless communication device, a remote device, a mobile subscriber station, an access terminal, May also be referred to as a mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

As used herein, the terms "determining" and "determining" may encompass a wide variety of operations. "Judgment" and "determination" are, for example, calculating, computing, processing, deriving, investigating, looking up (e.g., Searching in data structures), and regarding confirmation (ascertaining) as ``judgment'' or ``decision''. In addition, "judgment" and "decision" refer to receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, and access. (accessing) (e.g., accessing data in memory) may include considering something as a "judgment" or "decision." In addition, "judgment" and "decision" refer to resolving, selecting, choosing, establishing, comparing, etc. as "judgment" and "decision". may be included. In other words, "judgment" and "decision" may include regarding some action as having been "judged" or "determined."

As used herein, the phrase "based on" does not mean "based solely on" unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

Any reference to the elements herein, such as "first", "second", etc., does not generally limit the amount or order of those elements. These designations may be used herein as a convenient way of distinguishing between two or more elements. Thus, reference to a first and second element does not imply that only two elements may be employed therein or that the first element must precede the second element in any way.

To the extent that the words "include," "including," and variations thereof are used in this specification or in the claims, these terms are synonymous with the term "comprising." is intended to be comprehensive. Furthermore, the term "or" as used in this specification or in the claims is not intended to be exclusive or.

In this specification, a plurality of devices is also included unless it is clear from the context or technology that only one device exists.

Throughout this disclosure, the plural is intended to be included unless the context clearly dictates otherwise.

DESCRIPTION OF SYMBOLS 10... Dialogue device, 14... Score derivation part, 15... Wrong answer determination part, 16... Correct answer determination part, 17... Question sentence generation part, 18... Output part, 50... User terminal (user).

Claims (6)

an incorrect answer determining unit that determines an incorrect answer term that is similar in pronunciation to the input term that is a term input by the user or a term related to the term;
a correct answer determining unit that determines a correct answer term that is similar in meaning to the incorrect answer term;
a question sentence generation unit that generates a question sentence using the correct answer term as an answer;
an output unit that outputs the question text ,
The question sentence generation unit is an interactive device that extracts a sentence related to the correct answer term using a search function and generates the question sentence based on the sentence. For the plurality of incorrect answer term candidates , a score related to validity as an incorrect answer term is derived so that the higher the pronunciation similarity with the input term, the more likely it is to be selected . further comprising a score derivation unit that derives a score regarding validity as a correct answer term so that the higher the similarity in meaning with the answer term, the more likely it is to be selected ;
The incorrect answer determining unit determines an incorrect answer term from the plurality of incorrect answer term candidates based on the score derived by the score deriving unit,
The dialog device according to claim 1, wherein the correct answer determining unit determines a correct answer term from the plurality of correct answer term candidates based on the score derived by the score deriving unit. The dialog device according to claim 2, wherein the score deriving unit derives a score regarding the validity of each of the plurality of incorrect answer term candidates as the incorrect answer term by calculating a phonetic distance from the input term. . The dialog according to claim 2 or 3, wherein the score derivation unit derives a score regarding the validity of each of the plurality of correct term candidates as the correct term by calculating a semantic distance from the incorrect term. Device. The score derivation unit calculates the popularity of the term, and calculates a score regarding the validity of the term as the incorrect term, such that the higher the popularity of the term, the more likely it is to be selected for each of the plurality of incorrect term candidates. 5. At the same time as deriving the correct answer term, a score regarding validity as the correct answer term is derived for each of the plurality of correct answer term candidates such that the higher the popularity of the term, the more likely it is to be selected. The dialogue device according to item 1. 2. The dialogue device according to claim 1, wherein the output unit outputs the question sentence after outputting a repeated utterance guidance sentence for causing the input term to be repeatedly uttered.

Images