Automatic transaction apparatus, transaction control method, and transaction control program

Abstract

To improve the convenience of transactions by an automatic transaction apparatus.SOLUTION: An extraction part extracts text information from input voice. A specification part identifies multiple categories of transaction information based on the text information. An acquisition part outputs an input request for requesting the input of missing information not included in the text information among information of the multiple categories, and acquires the missing information according to the output of the input request. A generation part generates the transaction information including information of the multiple categories using the acquired missing information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an automated transaction device and a transaction control technology. [Background technology]

[0002] When users of ATMs (Automatic Teller Machines) at financial institutions and other institutions make transactions, they may enter information such as account numbers using a touch panel or by using a handset to input data via a VI (Voice Interface). When inputting data via a VI, users follow the voice guidance output from the handset and input numbers using the numeric keypad.

[0003] Regarding ATMs, a voice guidance processing device that allows smooth information input using a handset is known (see, for example, Patent Document 1). A customer service server that calculates a predicted end time that is close to the actual end time is also known (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-130455 [Patent Document 2] Japanese Patent Application Laid-Open No. 2011-65304 Summary of the Invention [Problem to be solved by the invention]

[0005] As ATMs become more common, there has been an increase in inquiries from users regarding ATM operation instructions. Visually impaired people also use handset-based VIs to conduct transactions instead of touch panel operations. Even visually impaired people will have to use handset-based VIs to conduct transactions if their touch panel breaks down.

[0006] However, there are currently limitations on the transactions that can be made using VI. For example, many financial institutions do not accept transfers using VI.

[0007] This problem is not limited to ATMs, but occurs in a variety of automated transaction devices.

[0008] In one aspect, the present invention aims to improve the convenience of transactions using an automated transaction device. [Means for solving the problem]

[0009] According to one embodiment, the automated transaction device includes an extracting unit, an identifying unit, an acquiring unit, and a generating unit.

[0010] The extraction unit extracts text information from the input speech. The identification unit identifies multiple categories of transaction information based on the text information. The acquisition unit outputs an input request requesting input of missing information not included in the text information from among the information in each of the multiple categories, and acquires the missing information in response to the output of the input request. The generation unit uses the acquired missing information to generate transaction information including information in each of the multiple categories.

[0011] According to another embodiment, a computer performs the following steps.

[0012] The computer extracts text information from the input speech. The computer identifies multiple categories of transaction information based on the text information. The computer outputs an input request requesting input of missing information not included in the text information from among the information in each of the multiple categories, and acquires the missing information in response to the output of the input request. The computer uses the acquired missing information to generate transaction information including information in each of the multiple categories.

[0013] According to yet another embodiment, the transaction control program causes a computer to perform the following process.

[0014] The computer extracts text information from the input speech. The computer identifies multiple categories of transaction information based on the text information. The computer outputs an input request requesting input of missing information not included in the text information from among the information in each of the multiple categories, and acquires the missing information in response to the output of the input request. The computer uses the acquired missing information to generate transaction information including information in each of the multiple categories. [Effects of the Invention]

[0015] According to one aspect, the convenience of transactions using an automated transaction device can be improved. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a functional configuration diagram of an automatic transaction device according to an embodiment. [Figure 2] 10 is a flowchart of a first transaction control process. [Figure 3] FIG. 1 is a diagram illustrating the configuration of a trading system. [Figure 4] FIG. 1 is a functional configuration diagram of an ATM. [Figure 5] FIG. 10 is a diagram showing a first keyword table. [Figure 6] FIG. 10 is a diagram showing a transaction name table. [Figure 7] FIG. 10 is a diagram showing a financial institution name table. [Figure 8] FIG. 10 is a diagram illustrating a branch name table. [Figure 9] FIG. 10 is a diagram illustrating a category table. [Figure 10] FIG. 10 is a diagram illustrating a branch name table. [Figure 11] FIG. 10 is a diagram illustrating a category table. [Figure 12A] 10 is a flowchart (part 1) of the second transaction control process. [Figure 12B] 10 is a flowchart (part 2) of the second transaction control process. [Figure 13] FIG. 10 is a diagram showing a second keyword table. [Figure 14] FIG. 10 is a diagram showing a third keyword table. [Figure 15] FIG. 2 is a hardware configuration diagram of an information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0017] Hereinafter, embodiments will be described in detail with reference to the drawings.

[0018] 1 shows an example of the functional configuration of an automated teller machine according to an embodiment. The automated teller machine 101 in FIG. 1 includes an extracting unit 111, an identifying unit 112, an acquiring unit 113, and a generating unit 114.

[0019] Fig. 2 is a flowchart showing an example of a first transaction control process performed by the automated transaction device 101 of Fig. 1. First, the extraction unit 111 extracts text information from the input voice (step 201), and the identification unit 112 identifies multiple categories of transaction information based on the text information (step 202).

[0020] Next, the acquisition unit 113 outputs an input request to request input of missing information that is not included in the text information among the information of each of the multiple categories (step 203), and acquires the missing information in response to the output of the input request (step 204).The generation unit 114 then uses the acquired missing information to generate transaction information that includes information of each of the multiple categories (step 205).

[0021] According to the automated teller machine 101 of FIG. 1, the convenience of transactions using the automated teller machine 101 can be improved.

[0022] The automated transaction device 101 is an ATM, currency exchange machine, teller machine, etc., and deposits and dispenses cash such as banknotes and coins managed by retail stores, department stores, supermarkets, etc. that are affiliated with banks and credit card companies. The cash is not limited to banknotes and coins, but may also be securities such as checks and bills, or media with monetary value such as tickets.

[0023] Fig. 3 shows an example of the configuration of a transaction system including the automated teller machine 101 of Fig. 1. The transaction system of Fig. 3 includes an ATM 311 and a host server 312. The ATM 311 is an example of the automated teller machine 101 of Fig. 1.

[0024] The ATM 311 and the host server 312 can communicate with each other via a communication network 313. Two or more ATMs 311 may be connected to the communication network 313.

[0025] Host server 312 is an information processing device (computer) of a financial institution, and manages information on cash exchanged in transactions using ATM 311. Transactions using ATM 311 include transfers, payments, and deposits. A financial institution is an organization that conducts business related to financial transactions, such as a bank, a credit union, or a credit card company.

[0026] Fig. 4 shows an example of the functional configuration of the ATM 311 in Fig. 3. The ATM 311 in Fig. 4 includes an input unit 411, a display unit 412, a handset 413, a reading unit 414, a communication unit 415, an extraction unit 416, an identification unit 417, an acquisition unit 418, a generation unit 419, a control unit 420, a cash storage unit 421, and a memory unit 422.

[0027] The input unit 411, the display unit 412, the handset 413, the reading unit 414, the communication unit 415, the cash storage unit 421, and the memory unit 422 are hardware. The extraction unit 416, the identification unit 417, the acquisition unit 418, and the generation unit 419 correspond to the extraction unit 111, the identification unit 112, the acquisition unit 113, and the generation unit 114 in FIG. 1, respectively.

[0028] The display unit 412 displays an operation screen including a transaction menu, text messages, etc. The input unit 411 accepts operation inputs from a user on the operation screen and generates operation signals indicating the accepted operation inputs. The combination of the input unit 411 and the display unit 412 is sometimes called a touch panel.

[0029] Handset 413 includes microphone 431, speaker 432, and numeric input device 433. Microphone 431 receives voice input from the user and outputs the received voice to extraction unit 416. Speaker 432 outputs a voice message to the user.

[0030] The numeric input device 433 includes a plurality of input buttons, accepts an operational input from the user, and outputs information such as a numeric value indicated by the accepted operational input to the acquisition unit 418. The numeric input device 433 is also called a numeric keypad.

[0031] When a user presents a recording medium such as a cash card or a passbook, the reading unit 414 reads information from the presented recording medium and outputs it to the acquiring unit 418. The cash card is a contact-type cash card such as a magnetic card, or a contactless-type cash card such as an IC (Integrated Circuit) card.

[0032] For example, if the cash card is a magnetic card, the user presents the cash card to the ATM 311 by inserting the cash card into the insertion slot of the reading unit 414. If the cash card is a contactless IC card, the user presents the cash card to the ATM 311 by bringing the cash card close to the reading unit 414. The user can also present a passbook or a transfer ticket to the ATM 311 by inserting the passbook or a transfer ticket into the insertion slot of the reading unit 414.

[0033] The communication unit 415 communicates with the host server 312 via the communication network 313 .

[0034] The storage unit 422 stores a keyword table 441 , a transaction name table 442 , a financial institution name table 443 , a branch name table 444 , a category table 445 , a message table 446 , and a recording medium table 447 .

[0035] Keyword table 441 includes keywords extracted from speech input through microphone 431 of handset 413 and information associated with the keywords.

[0036] The extraction unit 416 converts the input speech into text data and extracts nouns and verbs from the text data by morphological analysis. The extraction unit 416 then stores the extracted nouns and verbs as keywords in the keyword table 441. Keywords are an example of text information.

[0037] Fig. 5 shows an example of the first keyword table 441. The keyword table 441 in Fig. 5 includes an item number, a keyword, a category, and a content. The item number, keyword, category, and content correspond to one another. The item number is an identification number of an entry in the keyword table 441. The keyword represents a keyword extracted from the voice input by the microphone 431. The category represents the category of information corresponding to the keyword, and the content represents that information.

[0038] In this example, it is assumed that a voice message saying "Please request the transaction you wish to make, such as withdrawal, deposit, or transfer" is output from speaker 432, and a voice saying "I would like to transfer money to AAA Bank" is input through microphone 431.

[0039] In this case, "AAA" and "bank" are extracted as nouns from the input speech, and "transfer" is extracted as a verb. "AAA" is stored as a keyword with item number "1," "bank" is stored as a keyword with item number "2," and "transfer" is stored as a keyword with item number "3."

[0040] The transaction name table 442 includes each keyword and the transaction name associated with the keyword.

[0041] Fig. 6 shows an example of a transaction name table 442. The transaction name table 442 in Fig. 6 includes keywords, categories, and transaction names. The keywords, categories, and transaction names correspond to one another. The keywords represent keywords extracted from the voice input by the microphone 431. The categories represent the categories of the keywords, and the transaction names represent the names of the transactions corresponding to the keywords.

[0042] In this example, the categories of the keywords "withdrawal," "withdrawal," "transfer," "deposit," "deposit," and "deposit" are all "transaction names." The transaction name corresponding to "withdrawal" and "withdrawal" is "payment," the transaction name corresponding to "transfer" is "transfer," and the transaction name corresponding to "deposit," "deposit," and "deposit" is "deposit."

[0043] When a keyword W stored in the keyword table 441 is included in the transaction name table 442, the identification unit 417 acquires a transaction name corresponding to the keyword W from the transaction name table 442. Then, the identification unit 417 stores "transaction name" in the keyword table 441 as a category corresponding to the keyword W, and stores the acquired transaction name as content corresponding to "transaction name."

[0044] 5, since the keyword "transfer" is included in transaction name table 442, the transaction name "transfer" corresponding to the keyword "transfer" is obtained from transaction name table 442. Then, in keyword table 441, "transaction name" is stored as the category corresponding to the keyword "transfer", and the transaction name "transfer" is stored as the content corresponding to the category "transaction name".

[0045] The financial institution name table 443 includes the name of each financial institution and a financial institution code associated with the financial institution name.

[0046] Fig. 7 shows an example of financial institution name table 443. Financial institution name table 443 in Fig. 7 includes categories, financial institution names, and financial institution codes. The categories, financial institution names, and financial institution codes correspond to one another. The categories represent the categories of financial institution names, the financial institution names represent the names of financial institutions, and the financial institution codes represent identification information of financial institutions.

[0047] In this example, the categories of "AAA Bank," "BBB Bank," and "CCC Bank" are all "transfer destination financial institution name." The financial institution code corresponding to "AAA Bank" is "aaaa," the financial institution code corresponding to "BBB Bank" is "bbbb," and the financial institution code corresponding to "CCC Bank" is "cccc."

[0048] If keyword W stored in keyword table 441 is included in the name of a financial institution in financial institution name table 443, identification unit 417 acquires the name of the financial institution from financial institution name table 443. Then, identification unit 417 stores "transfer destination financial institution name" in keyword table 441 as the category corresponding to keyword W, and stores the acquired financial institution name as the content corresponding to "transfer destination financial institution name."

[0049] 5, since the keyword "AAA" is included in the financial institution name "AAA Bank" in financial institution name table 443, "AAA Bank" is obtained from financial institution name table 443. Then, in keyword table 441, "transfer destination financial institution name" is stored as the category corresponding to the keyword "AAA", and the financial institution name "AAA Bank" is stored as the content corresponding to the category "transfer destination financial institution name".

[0050] The branch name table 444 includes the branch name of each financial institution and the branch code associated with the branch name.

[0051] Figure 8 shows an example of branch name table 444. Branch name table 444 in Figure 8 includes financial institution names, branch names, kana pronunciations of branch names, and branch codes. Financial institution names, branch names, kana pronunciations of branch names, and branch codes are associated with each other. Financial institution names represent financial institution names included in financial institution name table 443, branch names represent the names of each branch of a financial institution, branch name pronunciations represent the pronunciations of branch names, and branch codes represent identification information of branches.

[0052] In this example, the financial institution names for "Head Office Sales Department," "XX Branch," and "XX Branch" are all "AAA Bank." The branch name pronunciation corresponding to "Head Office Sales Department" is "Honten," and the branch code is "100." The branch name pronunciation corresponding to "XX Branch" is "Batsubatsu," and the branch code is "101." The branch name pronunciation corresponding to "XX Branch" is "Marumaru," and the branch code is "102."

[0053] The category table 445 includes the transaction name of each transaction and the category associated with the transaction name.

[0054] Fig. 9 shows an example of category table 445. Category table 445 in Fig. 9 includes transaction names and categories C1 to C10. Transaction names and categories C1 to C10 are associated with each other. Transaction names represent transaction names included in transaction name table 442, and categories C1 to C10 represent categories of information required for the transaction indicated by the transaction names.

[0055] In this example, category C1 corresponding to "payment" is "withdrawal amount," category C2 is "PIN," category C3 is "user financial institution code," category C4 is "user branch code," and category C5 is "user account number."

[0056] Category C1 corresponding to "transfer" is "transfer method," category C2 is "recipient financial institution name," category C3 is "recipient branch name," and category C4 is "transfer amount." Category C5 corresponding to "transfer" is "recipient account item," and category C6 is "recipient account number."

[0057] Category C7 corresponding to "transfer" is "PIN number", category C8 is "user financial institution code", category C9 is "user branch code", and category C10 is "user account number".

[0058] Category C1 corresponding to "deposit" is "deposit amount", category C2 is "user financial institution code", category C3 is "user branch code", and category C4 is "user account number".

[0059] Categories C1 to C10 in category table 445 are associated via transaction names with any of the keywords in transaction name table 442. Transaction name table 442 and category table 445 are an example of category-specific information that includes a plurality of keywords and a plurality of categories related to transactions that are associated with each of the plurality of keywords.

[0060] The message table 446 includes each category and a text message associated with the category.

[0061] Fig. 10 shows an example of message table 446. Message table 446 in Fig. 10 includes categories, reply methods, and messages. Categories, reply methods, and messages correspond to one another. Categories represent categories included in category table 445. Reply methods represent the way in which a user will reply to a question inquiring about the content of a category, and messages represent the content of the question.

[0062] In this example, the answering method is "voice" or "numeric input", and the message is output as a voice message from speaker 432 of handset 413. If the answering method is "numeric input", the message contains the character string "Please enter using the numeric keypad on your handset."

[0063] When the answering method is "voice," the user inputs a voice indicating the answer using microphone 431 of handset 413. By obtaining the answer by voice, the user can input the answer without using a touch panel.

[0064] When the answering method is "numerical input," the user inputs a numerical value indicating the answer using the numerical input device 433 of the handset 413. By inputting the answer numerically rather than by voice, the security of information that you do not want others to hear, such as a transfer account number or PIN, can be ensured.

[0065] The recording medium table 447 includes the transaction name of each transaction and information about the recording medium associated with the transaction name.

[0066] Fig. 11 shows an example of the recording medium table 447. The recording medium table 447 in Fig. 11 includes the transaction name, transfer method, medium M1 to medium M3, necessity K1 to necessity K3, and message. The transaction name, transfer method, medium M1 to medium M3, necessity K1 to necessity K3, and message are associated with each other.

[0067] The transaction name indicates the transaction name included in the transaction name table 442, and the transfer method indicates the transfer method when the transaction name is "transfer." Medium M1 to medium M3 indicate the types of recording media that can be used, and necessity Ki (i = 1 to 3) indicates the necessity of medium Mi. The message indicates a message requesting the user to present the recording media corresponding to the transaction name. The message is output as a voice message from speaker 432 of handset 413.

[0068] In this example, the transfer method is "bank transfer" or "cash transfer." The medium M1 corresponding to "payment" is "card," and necessity K1 is "required," while the medium M2 is "passbook," and necessity K2 is "optional." "Card" represents a cash card.

[0069] The medium M1 corresponding to "account transfer" is a "card" and necessity K1 is "required", the medium M2 is a "passbook" and necessity K2 is "optional", and the medium M3 is a "transfer ticket" and necessity K3 is "optional".

[0070] The medium M1 corresponding to "cash transfer" is "cash" and necessity K1 is "essential", and the medium M2 is a "transfer slip" and necessity K2 is "optional".

[0071] The medium M1 corresponding to "deposit" is a "card" and necessity K1 is "either one is required," the medium M2 is a "passbook" and necessity K2 is "either one is required," and the medium M3 is "cash" and necessity K3 is "required." In this case, necessity K1 and necessity K2 indicate that either medium M1 or medium M2 is required.

[0072] The identification unit 417 acquires the contents of the categories "user financial institution code," "user branch code," and "user account number" included in the category table 445 from the user's cash card, passbook, transfer slip, etc.

[0073] First, the identification unit 417 obtains from the recording medium table 447 a message requesting the user to present a recording medium corresponding to the contents of the category "transaction name" in the keyword table 441. The identification unit 417 then outputs the obtained message as a voice message via the speaker 432 of the handset 413, and obtains the financial institution code, branch code, and account number from the information recorded on the presented recording medium.

[0074] Next, the identification unit 417 stores the "user financial institution code," "user branch code," and "user account number" as categories in the keyword table 441. Then, the identification unit 417 stores the acquired financial institution code, branch code, and account number as the contents of the "user financial institution code," "user branch code," and "user account number," respectively.

[0075] Next, the identification unit 417 acquires from the category table 445 multiple categories corresponding to the contents of the category "transaction name" in the keyword table 441, and identifies the acquired categories as multiple categories of the transaction information. Then, the identification unit 417 stores in the keyword table 441 the identified categories other than the categories already stored in the keyword table 441.

[0076] Acquisition unit 418 acquires from message table 446 a message inquiring of the user about the content of a category whose content is not stored among the multiple categories stored in keyword table 441. Acquisition unit 418 then outputs the acquired message as a voice message via speaker 432 of handset 413, and acquires a response from the user in response to the output of the voice message.

[0077] Content that is not stored in keyword table 441 is an example of missing information. A voice message inquiring of the user about content that is not stored in keyword table 441 is an example of an input request that requests input of missing information. A voice message that includes a request to "enter using the numeric keypad on the handset" is an example of an input request that requests input of missing information using numeric input device 433.

[0078] When the user inputs a voice indicating an answer using microphone 431 of handset 413, extraction unit 416 converts the input voice into text data and extracts nouns from the text data by morphological analysis. Extraction unit 416 then outputs the extracted nouns to acquisition unit 418, which stores the nouns in keyword table 441 as the content of the inquiry. The nouns extracted from the text data are an example of additional text information.

[0079] When the user inputs a number indicating an answer using the number input device 433 of the handset 413, the acquisition unit 418 stores the information indicated by the input number in the keyword table 441 as the content of the inquiry.

[0080] Generation unit 419 generates transaction information including the contents of each category stored in keyword table 441 , and transmits the generated transaction information to host server 312 via communication unit 415 .

[0081] The cash storage unit 421 stores cash managed by a financial institution. The control unit 420 controls the overall operation of the ATM 311 including the cash storage unit 421.

[0082] Figures 12A and 12B are flowcharts showing an example of the second transaction control process performed by the ATM 311 of Figure 4. First, the microphone 431 of the handset 413 receives voice input from the user (step 1201).

[0083] Next, the extraction unit 416 converts the received speech into text data (step 1202), and extracts nouns and verbs from the text data by morphological analysis (step 1203).Then, the extraction unit 416 stores the extracted nouns and verbs as keywords in the keyword table 441 (step 1204).

[0084] In the example of Figure 5, "AAA," "bank," and "transfer" are extracted from the speech "I would like to transfer money to AAA Bank," and are stored as keywords in item numbers "1" to "3." Of these, "AAA" is included in the financial institution name "AAA Bank" in financial institution name table 443 of Figure 7, so "recipient financial institution name" is stored as the category corresponding to the keyword "AAA," and the financial institution name "AAA Bank" is stored as the content corresponding to the category "recipient financial institution name."

[0085] Next, the identification unit 417 refers to the transaction name table 442, and if any of the keywords stored in the keyword table 441 is included in the transaction name table 442, it acquires the transaction name corresponding to that keyword. Then, the identification unit 417 stores "transaction name" in the keyword table 441 as the category corresponding to that keyword, and stores the acquired transaction name as the content corresponding to "transaction name" (step 1205).

[0086] In the example of Figure 5, since the keyword "transfer" is included in the transaction name table 442 of Figure 6, "transaction name" is stored as the category corresponding to the keyword "transfer", and the transaction name "transfer" is stored as the content corresponding to the category "transaction name".

[0087] If none of the keywords is included in the trading name table 442, the processing of step 1205 is skipped.

[0088] Next, the identifying unit 417 checks whether the category "transaction name" is stored in the keyword table 441 (step 1206).

[0089] If the category "transaction name" is not stored in keyword table 441 (step 1206, NO), identification unit 417 refers to message table 446 and acquires a message corresponding to the category "transaction name." Then, identification unit 417 outputs the acquired message as a voice message via speaker 432 of handset 413 (step 1212).

[0090] 10, the message corresponding to the category "transaction name" does not include the request "Please enter using the numeric keypad on the handset," so the user responds by voice. Then, the ATM 311 repeats the process from step 1201 onwards.

[0091] If the category "transaction name" is stored in the keyword table 441 (step 1206, YES), the identifying unit 417 checks the contents of the category "transaction name" (step 1207).

[0092] If the content of the category "transaction name" is "transfer" (step 1207, YES), the identifying unit 417 acquires the transfer method from the user and stores it in the keyword table 441 (step 1208).

[0093] In step 1208, the identification unit 417 refers to the message table 446 to acquire a message corresponding to the category "transfer method." The identification unit 417 then outputs the acquired message as a voice message via the speaker 432.

[0094] In message table 446 of FIG. 10, the message corresponding to the category "transfer method" does not include the request "Please enter using the numeric keypad on the handset.", so the user responds by voice.

[0095] Therefore, extraction unit 416 converts the input speech into text data, extracts nouns from the text data using morphological analysis, and outputs them to identification unit 417. Identification unit 417 stores "transfer method" as a category in keyword table 441, and stores the extracted nouns as the content of "transfer method."

[0096] In the example of Fig. 5, since the content of the category "transaction name" is "transfer," a voice message corresponding to the category "transfer method" is output in message table 446 of Fig. 10. Then, "bank transfer" is stored as the content of the category "transfer method."

[0097] Next, the identification unit 417 requests the user to present the recording medium (step 1209). In step 1209, the identification unit 417 refers to the recording medium table 447 and acquires a message corresponding to the content of the category "transaction name" in the keyword table 441. If the content of the category "transaction name" is "transfer", the identification unit 417 acquires a message corresponding to the content of the category "transfer method" in the keyword table 441. The identification unit 417 then outputs the acquired message as a voice message via the speaker 432.

[0098] Next, the identification unit 417 acquires the financial institution code, branch code, and account number from the information recorded on the presented recording medium, and stores them in the keyword table 441 (step 1210).

[0099] In step 1210, the identification unit 417 stores the "user financial institution code," "user branch code," and "user account number" as categories in the keyword table 441. The identification unit 417 then stores the acquired financial institution code, branch code, and account number as the contents of the "user financial institution code," "user branch code," and "user account number," respectively.

[0100] In the example of Fig. 5, since the content of the category "Transfer Method" is "Account Transfer," a voice message corresponding to "Account Transfer" is output in recording medium table 447 of Fig. 11. Then, "aaaa" is stored as the content of "User Financial Institution Code," "100" is stored as the content of "User Branch Code," and "7654321" is stored as the content of "User Account Number."

[0101] If the "transfer method" is "cash transfer", the "user financial institution code", "user branch code", and "user account number" are not stored in the keyword table 441.

[0102] Next, the identification unit 417 refers to the category table 445, acquires multiple categories corresponding to the contents of the category "transaction name" in the keyword table 441, and identifies the acquired categories as multiple categories of the transaction information. Then, the identification unit 417 stores, in the keyword table 441, the identified categories other than the categories already stored in the keyword table 441 (step 1211).

[0103] In the example of FIG. 5, information on categories C1 to C10 is acquired from category table 445 as categories corresponding to the content "transfer" of the category "transaction name."

[0104] Of the information in categories C1 to C11, "transfer method," "recipient financial institution name," "user financial institution code," "user branch code," and "user account number" are already stored in keyword table 441. Therefore, the remaining "recipient branch name," "transfer amount," "recipient account," "recipient account number," and "PIN number" are stored in keyword table 441.

[0105] In this way, by using the transaction name table 442 and the category table 445, the category of information required for the transaction can be identified from the contents of the "transaction name," and the category of missing information can be extracted from the identified categories.

[0106] If the content of the category "transaction name" is not "transfer" (step 1207, NO), the ATM 311 performs the process from step 1209 onwards.

[0107] Next, the acquisition unit 418 checks whether the contents of all categories in the keyword table 441 are stored (step 1213).

[0108] If the contents of any category are not stored (step 1213, NO), the acquisition unit 418 selects a category whose contents are not stored (step 1221), and acquires a message corresponding to the contents from the message table 446. The acquisition unit 418 then outputs the acquired message as a voice message via the speaker 432, thereby inquiring the user about the contents of the category (step 1222).

[0109] Next, the acquisition unit 418 checks the answering method for the selected category by referring to the message table 446 (step 1223). If the answering method is "voice" (step 1223, YES), the acquisition unit 418 requests the extraction unit 416 to perform voice recognition.

[0110] The extraction unit 416 converts the voice received by the microphone 431 into text data (step 1224), and extracts nouns from the text data by morphological analysis (step 1225). Then, the extraction unit 416 outputs the extracted nouns to the acquisition unit 418.

[0111] If the noun output from the extraction unit 416 matches the queried content, the acquisition unit 418 stores the noun as the content of the selected category in the keyword table 441 (step 1226). Then, the ATM 311 repeats the processes from step 1213 onwards. If the output noun does not match the queried content, the ATM 311 repeats the same processes as steps 1222 to 1225.

[0112] If the answer method is "numerical input" (step 1223, NO), the acquisition unit 418 stores the information indicated by the numerical value input by the numerical value input device 433 of the handset 413 as the content of the selected category in the keyword table 441 (step 1227). Then, the ATM 311 repeats the processing from step 1213 onwards.

[0113] In the message table 446 of Fig. 10, the response method for "Name of transfer branch" is "voice." Therefore, in the example of Fig. 5, "Head Office Sales Department" is acquired by voice and stored as the content of "Name of transfer branch."

[0114] In addition, in the message table 446 of Fig. 10, the answering method for "transfer amount," "transfer destination account," "transfer destination account number," and "PIN" is "numerical input." Therefore, in the example of Fig. 5, "10000," "regular," "1234567," and "1234" are obtained by numerical input and stored as the contents of "transfer amount," "transfer destination account," "transfer destination account number," and "PIN," respectively.

[0115] If the contents of all categories are stored in keyword table 441 (YES at step 1213 ), acquisition unit 418 performs the process of step 1214 .

[0116] In step 1214, the acquisition unit 418 refers to the financial institution name table 443 and acquires the financial institution code corresponding to the content of the category "transfer destination financial institution name" in the keyword table 441. The acquisition unit 418 then stores "transfer destination financial institution code" as a category in the keyword table 441, and stores the acquired financial institution code as the content corresponding to "transfer destination financial institution code."

[0117] In the example of Figure 5, since the content of the category "recipient financial institution name" is "AAA Bank," "aaaa" is obtained as the financial institution code corresponding to the financial institution name "AAA Bank" from financial institution name table 443 in Figure 7. Then, the financial institution code "aaaa" is stored as the content corresponding to the category "recipient financial institution code."

[0118] Next, the acquisition unit 418 refers to the branch name table 444 and acquires the branch code corresponding to the content of the category "transfer destination branch name" in the keyword table 441. The acquisition unit 418 then stores "transfer destination branch code" as a category in the keyword table 441, and stores the acquired branch code as the content corresponding to "transfer destination branch code" (step 1215).

[0119] In the example of Figure 5, since the content of the category "transfer destination branch name" is "head office sales department", "100" is obtained as the branch code corresponding to the branch name "head office sales department" from the branch name table 444 in Figure 8. Then, the branch code "100" is stored as the content corresponding to the category "transfer destination branch code".

[0120] If "recipient financial institution name" is not stored as a category in keyword table 441, then "recipient financial institution code" is not stored in keyword table 441 in step 1214. If "recipient branch name" is not stored as a category in keyword table 441, then "recipient branch code" is not stored in keyword table 441 in step 1215.

[0121] Next, the acquisition unit 418 checks the content of the category "transaction name" (step 1216). If the content of the category "transaction name" is "transfer" (step 1216, YES), the acquisition unit 418 performs the process of step 1217.

[0122] In step 1217, the acquisition unit 418 generates inquiry information including the contents of the “transfer destination financial institution code,” “transfer destination subject,” and “transfer destination account number” in the keyword table 441.

[0123] Next, acquisition unit 418 sends the generated inquiry information to host server 312 via communication unit 415, and receives the account holder name of the transfer destination account indicated in the inquiry information from host server 312. Acquisition unit 418 then stores "transfer destination account holder name" as a category in keyword table 441, and stores the received account holder name as content corresponding to "transfer destination account holder name."

[0124] In the example of Figure 5, "Yamada Taro" is acquired as the account holder name of the transfer destination account indicated by the contents of "transfer destination financial institution code," "transfer destination account," and "transfer destination account number." The account holder name "Yamada Taro" is then stored as the content corresponding to the category "transfer destination account holder name."

[0125] Next, the acquisition unit 418 outputs the content of each category in the keyword table 441 via the display unit 412 or the speaker 432 (step 1218). Then, the user checks the output content and uses the input unit 411 or the microphone 431 to input confirmation information indicating that the user has checked the content.

[0126] Next, generation unit 419 generates transaction information including the contents of each category in keyword table 441, and transmits the generated transaction information to host server 312 via communication unit 415 (step 1219).

[0127] Next, the control unit 420 notifies the user of the transaction result via the display unit 412 or the speaker 432, and outputs a transaction statement (step 1220). If a recording medium is inserted, the control unit 420 ejects the inserted recording medium. If the transaction name is "Payment", the control unit 420 takes out cash from the cash storage unit 421 and ejects it.

[0128] If the content of the category "transaction name" is not "transfer" (step 1216, NO), the ATM 311 performs the process from step 1218 onwards.

[0129] 13 shows an example of the second keyword table 441. In this example, it is assumed that a voice message saying "Please request your desired transaction, such as withdrawal, deposit, or transfer" is output from the speaker 432, and a voice saying "I would like to deposit money into my account" is input through the microphone 431.

[0130] In this case, "account" is extracted as a noun and "deposit" is extracted as a verb from the input speech. "Account" is stored as a keyword with item number "1," and "deposit" is stored as a keyword with item number "2."

[0131] 6, the keyword "deposit" is included in the transaction name table 442, so the transaction name "deposit" corresponding to the keyword "deposit" is obtained from the transaction name table 442. Then, in the keyword table 441, "transaction name" is stored as the category corresponding to the keyword "deposit", and the transaction name "deposit" is stored as the content corresponding to the category "transaction name".

[0132] Next, a voice message corresponding to "Deposit" is output in recording medium table 447 of Fig. 11. Then, "bbbb" is stored as the content of "User financial institution code," "010" is stored as the content of "User branch code," and "0000123" is stored as the content of "User account number."

[0133] Next, information on categories C1 to C4 corresponding to the content "Deposit" of the category "Transaction Name" is obtained from category table 445 in Figure 9. Of these, "User Financial Institution Code," "User Branch Code," and "User Account Number" are already stored in keyword table 441. Therefore, the remaining "Deposit Amount" is stored in keyword table 441.

[0134] 10, the reply method for "deposit amount" is "numeric input." Therefore, "20000" is acquired by numeric input and stored as the content of "deposit amount."

[0135] 14 shows an example of the third keyword table 441. In this example, it is assumed that a voice message saying "Please request your desired transaction, such as withdrawal, deposit, or transfer" is output from the speaker 432 and a voice saying "I would like to withdraw money" is input through the microphone 431.

[0136] In this case, "money" is extracted as a noun and "drawer" is extracted as a verb from the input speech. "Money" is stored as a keyword with item number "1," and "drawer" is stored as a keyword with item number "2."

[0137] 6, the keyword "withdrawal" is included in the transaction name table 442, so the transaction name "payment" corresponding to the keyword "withdrawal" is obtained from the transaction name table 442. Then, in the keyword table 441, "transaction name" is stored as the category corresponding to the keyword "withdrawal", and the transaction name "payment" is stored as the content corresponding to the category "transaction name".

[0138] Next, a voice message corresponding to "Payment" is output in recording medium table 447 of Fig. 11. Then, "bbbb" is stored as the content of "User financial institution code," "010" is stored as the content of "User branch code," and "0000123" is stored as the content of "User account number."

[0139] Next, information on categories C1 to C5 corresponding to the content "payment" of the category "transaction name" is obtained from category table 445 in Figure 9. Of these, "user financial institution code," "user branch code," and "user account number" are already stored in keyword table 441. Therefore, the remaining "withdrawal amount" and "PIN number" are stored in keyword table 441.

[0140] 10, the answer method for "withdrawal amount" and "PIN number" is "numeric input." Therefore, "50000" and "1234" are acquired by numerical input and stored as the contents of "withdrawal amount" and "PIN number," respectively.

[0141] According to the transaction system of Fig. 3, missing information for a transaction is sequentially added through interactive operation using the handset 413, so that a transaction at the ATM 311 can be completed without the user having to operate the touch panel. Therefore, even if a visually impaired person is conducting a transaction or if the touch panel is broken, various transactions, including money transfers, can be performed using the handset 413. As such, there are no restrictions on the types of transactions, improving the convenience of transactions using the ATM 311.

[0142] The configuration of the automated teller machine 101 in FIG. 1 is merely an example, and some of the components may be omitted or changed depending on the application or conditions of the automated teller machine 101.

[0143] 3 is merely an example, and some components may be omitted or changed depending on the purpose or conditions of the transaction system. For example, the transaction system may include another automated transaction device such as a currency exchange machine or a teller machine instead of the ATM 311.

[0144] The configuration of the ATM 311 in Fig. 4 is merely an example, and some components may be omitted or changed depending on the application or conditions of the transaction system. As the configuration of another automated transaction device such as a currency exchange machine or a teller machine, for example, a configuration similar to that in Fig. 4 can be used.

[0145] The flowcharts of Figures 2, 12A, and 12B are merely examples, and some of the processes may be omitted or changed depending on the configuration or conditions of the trading system.

[0146] 5 to 11, 13, and 14 are merely examples, and tables of other formats may be used depending on the use or conditions of the ATM 311. For example, the ATM 311 may combine multiple tables and store them as a single table.

[0147] Fig. 15 shows an example of the hardware configuration of an information processing device used as the automated teller machine 101 in Fig. 1 and the ATM 311 in Fig. 3. The information processing device in Fig. 15 includes a CPU (Central Processing Unit) 1501, a memory 1502, an input device 1503, an output device 1504, an auxiliary storage device 1505, a media drive device 1506, and a network connection device 1507. These components are hardware and are connected to each other by a bus 1508.

[0148] The handset 413, reader 414, and cash storage unit 421 in FIG. 4 may be connected to a bus 1508.

[0149] The memory 1502 is, for example, a semiconductor memory such as a read-only memory (ROM) or a random access memory (RAM), and stores programs and data used in processing. The memory 1502 may operate as the storage unit 422 in FIG.

[0150] 1 by executing a program using memory 1502. CPU 1501 (processor) also operates as extraction unit 111, identification unit 112, acquisition unit 113, and generation unit 114 in FIG. 4 by executing a program using memory 1502.

[0151] The input device 1503 is, for example, a keyboard, a pointing device, etc., and is used for inputting instructions or information from the user. The output device 1504 is, for example, a display device, a printer, etc., and is used for making inquiries or instructions to the user and outputting processing results. The input device 1503 may operate as the input unit 411 in FIG. 4, and the output device 1504 may operate as the display unit 412 in FIG. 4. The input device 1503 and the output device 1504 may be touch panels. The processing result may be a transaction result.

[0152] The auxiliary storage device 1505 is, for example, a magnetic disk device, an optical disk device, a magneto-optical disk device, a tape device, or the like. The auxiliary storage device 1505 may be a hard disk drive or a solid state drive (SSD). The information processing device stores programs and data in the auxiliary storage device 1505 and can use them by loading them into the memory 1502. The auxiliary storage device 1505 may operate as the storage unit 422 in FIG. 4.

[0153] The medium drive device 1506 drives the portable recording medium 1509 and accesses the recorded contents thereof. The portable recording medium 1509 is a memory device, a flexible disk, an optical disk, a magneto-optical disk, etc. The portable recording medium 1509 may be a CD-ROM (Compact Disk Read Only Memory), a DVD (Digital Versatile Disk), a USB (Universal Serial Bus) memory, etc. An operator can store programs and data in the portable recording medium 1509 and load them into the memory 1502 for use.

[0154] In this way, the computer-readable recording medium that stores the program and data used in the processing is a physical (non-transitory) recording medium such as the memory 1502, the auxiliary storage device 1505, or the portable recording medium 1509.

[0155] The network connection device 1507 is a communication interface circuit that is connected to the communication network 313 and performs data conversion associated with communication. The information processing device receives programs and data from external devices via the network connection device 1507 and can load them into the memory 1502 for use. The network connection device 1507 may operate as the communication unit 415 in FIG. 4.

[0156] The host server 312 in FIG. 3 can be an information processing device similar to that in FIG.

[0157] 15, some components may be omitted or changed depending on the application or conditions of the information processing device. For example, if the portable recording medium 1509 is not used, the medium drive device 1506 can be omitted.

[0158] Although the disclosed embodiments and their advantages have been described in detail, those skilled in the art may make various modifications, additions, and omissions without departing from the scope of the invention as clearly set forth in the claims. [Explanation of symbols]

[0159] 101 Automatic transaction equipment 111, 416 Extraction part 112, 417 Specific part 113, 418 Acquisition Department 114, 419 generation part 311 ATM 312 Host Server 313 Communication Network 411 Input section 412 Display section 413 Handset 414 Reading unit 415 Communications Department 420 Control Unit 421 Cash Storage 422 Storage section 431 Mike 432 Speaker 433 Numerical Input Devices 441 Keyword Table 442 Transaction Name Table 443 Financial Institution Name Table 444 Branch Name Table 445 Category Table 446 Message Table 447 Recording Media Table 1501 CPU 1502 memory 1503 Input Device 1504 Output Device 1505 Auxiliary storage device 1506 Media drive unit 1507 Network connection device 1508 Bus 1509 Portable recording media

Claims

1. an extraction unit that extracts text information from input speech; an identification unit that identifies a plurality of categories of transaction information based on the text information; an acquisition unit that outputs an input request to request input of missing information that is not included in the text information among the information of each of the plurality of categories, and acquires the missing information in response to the output of the input request; a generation unit that generates the transaction information including information on each of the plurality of categories using the acquired missing information; An automatic transaction device comprising:

2. The system further includes a storage unit configured to store category-specific information including a plurality of keywords and a plurality of categories related to transactions, each of which is associated with one of the plurality of keywords; The automatic transaction device according to claim 1, characterized in that the identification unit identifies, as the multiple categories of the transaction information, multiple categories related to transactions associated with keywords included in the text information among the multiple keywords.

3. the acquisition unit outputs a voice message inquiring about the missing information as the input request; the extraction unit extracts additional text information from the input voice in response to the output of the voice message; 3. The automated transaction device according to claim 1, wherein the acquisition unit acquires the missing information from the additional text information.

4. The automatic transaction device described in claim 1 or 2, characterized in that the acquisition unit outputs a voice message requesting input of the missing information using a numerical input device as the input request, and acquires the missing information input from the numerical input device in response to the output of the voice message.

5. Extract text information from the input voice, identifying a plurality of categories of transaction information based on the text information; outputting an input request for requesting input of missing information not included in the text information among the information of each of the plurality of categories; acquiring the missing information in response to the output of the input request; generating the transaction information including information for each of the plurality of categories using the acquired missing information; A transaction control method characterized in that processing is executed by a computer.

6. Extract text information from the input voice, identifying a plurality of categories of transaction information based on the text information; outputting an input request for requesting input of missing information not included in the text information among the information of each of the plurality of categories; acquiring the missing information in response to the output of the input request; generating the transaction information including information for each of the plurality of categories using the acquired missing information; A transaction control program that causes a computer to execute processing.

Images