JPH0962869A - Medium issuing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the medium issuing device of a season ticket, which realizes a real time processing by means of a simple operation and which does not need operation skill. SOLUTION: A clerk in charge receives a season ticket purchase application P indicated from a customer. Then, the content mentioned of the received application P is inputted and the desired season ticket C is printed/issued by the medium issuing device. The clerk in charge, who receives the application P, interactively recognizes significant clauses (a train section, a use start date, the term of validity and the like) in the content mentioned of the application P to the customer, and sound information of 'from Kawasaki station to Shibuya station ?', which the clerk in charge utters, is inputted from a microphone 11. The inputted sound information is recognized by a sound recognition part 12. Specified words of 'from station' and 'to station' are retrieved by word spotting in a word recognition part 14 as against the recognized result, and the previous words of 'Kawasaki' and 'Shibuya' are recognized as the significant words of station names. Then, significant word information are printed in an issuing part 20 and they are issued as the season ticket C.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a medium issuing device for issuing a medium such as a commuter pass or a ticket used in transportation or a bankbook or a credit card used in a financial institution.

[0002]

2. Description of the Related Art For example, in a window service at a station or a bank of a transportation facility, a staff member at the window receives an application form that describes the necessary items presented by the customer, and based on the content of the application form,
A staff member prints and issues a commuter pass, a ticket, a passbook, a credit card, or the like by operating a keyboard of a terminal device, and provides it to a customer.

Here, taking the issue of a commuter pass as an example, the staff at the station window inputs the issue condition information such as the use start date, the use period, the use section, etc. described in the application form presented by the customer using the keyboard. By doing so, the commuter pass is printed and issued to the customer.

[0004]

However, conventionally, the real-time input means for the window service is only the keyboard method, and in order to issue it to the customer in a short time through complicated keyboard operation, it is necessary to use the operation method. There was a problem that it required skill. Therefore, it is an object of the present invention to provide a medium issuing device that enables real-time processing by a simple operation and does not require skill in operation.

[0005]

According to the medium issuing device of the present invention, a staff member receives an application form in which medium issuing condition information is entered, which is presented by a customer, and the medium issuing condition described in the received application form. A medium issuing device for issuing a medium according to the medium issuance condition information by inputting information, wherein a staff member who has received the application form makes a meaningful clause in the medium issuance condition information described in the application form. When interactively confirming with the customer, voice input means for inputting voice information uttered by an attendant, and voice processing means for detecting a significant word by recognizing the voice information input by the voice input means,
And an issuing means for issuing a predetermined medium based on the processing result of the audio processing means.

Further, in the medium issuing device of the present invention, the staff member receives the application form in which the medium issuing condition information is entered, which is presented by the customer, and inputs the medium issuing condition information described in the received application form. Thus, in the medium issuing device that issues the medium according to the medium issuing condition information, the staff member who receives the application form interactively outputs the meaningful clauses in the medium issuing condition information described in the application form. When confirming with the customer, voice input means for inputting voice information uttered by an attendant, voice recognition means for recognizing the voice information input by the voice input means, and a significant word for the recognition result of the voice recognition means It comprises a word recognizing means for recognizing a significant word by searching for a specific part of speech or a phrase following the, and an issuing means for issuing a predetermined medium based on the recognition result of the word recognizing means.

Further, the medium issuing device of the present invention is such that the customer manually inputs a part of the information displayed on the medium, and the clerk and the customer based on the information input by the manual input means. Voice input means for inputting voice information of a question and answer performed between the voice input means, voice processing means for detecting a significant word from semantic analysis of utterance by recognizing the voice information input by the voice input means, and the voice processing means. And a issuing unit for issuing a predetermined medium based on a combination of the processing result of the processing unit and the information input by the manual input unit.

Further, the medium issuing device of the present invention is such that the customer manually inputs a part of the information displayed on the medium, and the clerk and the customer based on the information input by the manual input means. And voice recognition means for recognizing the voice information input by the voice input means, and the voice recognition means for recognizing the voice information input by the voice input means. Word recognition means for recognizing a customer's answer and recognizing a significant word by searching for a specific part-of-speech or phrase following the significant word, and a set of the recognition result of this word recognition means and the information input by the manual input means. And a issuing means for issuing a predetermined medium based on the combination.

Further, the medium issuing device of the present invention recognizes the tablet-type manual input means by which the customer manually inputs a part of the character information displayed on the medium, and the character information input by the manual input means. Character recognition means, a voice input means for inputting voice information of a question and answer performed between the staff and the customer based on the character information input by the manual input means, and a voice input by the voice input means. Voice processing means for detecting a significant word from semantic analysis of utterance by recognizing information, and issuing means for issuing a predetermined medium based on a combination of the processing result of the voice processing means and the recognition result of the character recognition means. It is equipped with.

Further, the medium issuing device of the present invention recognizes the tablet-type manual input means by which the customer manually inputs a part of the character information displayed on the medium, and the character information input by the manual input means. Character recognition means, a voice input means for inputting voice information of a question and answer performed between the staff and the customer based on the character information input by the manual input means, and a voice input by the voice input means. For the voice recognition means that recognizes information and the recognition result of this voice recognition means,
Word recognition means for recognizing a customer's answer to a staff member's question and recognizing a significant word by searching for a specific part of speech or phrase following the significant word, and a recognition result of this word recognition means and a recognition result of the character recognition means And a issuing means for issuing a predetermined medium based on the combination of.

Further, the medium issuing device of the present invention is such that the customer manually inputs a part of the information displayed on the medium, and a voice for the customer to input the voice information according to a predetermined procedure. An input unit, a voice processing unit for detecting a significant word by recognizing voice information input by the voice input unit, and a combination of the processing result of the voice processing unit and the information input by the manual input unit. And a issuing means for issuing a predetermined medium on the basis.

Further, the medium issuing device of the present invention recognizes the tablet-type manual input means by which the customer manually inputs a part of the character information displayed on the medium, and the character information input by the manual input means. Character recognition means, a voice input means for the customer to input voice information according to a predetermined procedure, and a voice processing means for detecting a significant word by recognizing the voice information input by the voice input means. And an issuing means for issuing a predetermined medium based on a combination of the processing result of the voice processing means and the recognition result of the character recognition means.

According to the above-mentioned means, for example, by inputting and recognizing a natural conversation voice of the staff member with respect to the customer's request in the reception service, real-time processing by a simple operation becomes possible. You don't need to be skilled.

Furthermore, since the specific word of the specific speaker of the staff member is the recognition target, the efficiency of information input can be improved. Also,
For example, when issuing a medium such as a commuter pass or a personal card in the window reception work, by interactively exchanging information between the clerk and the customer, that is, by inputting text information using a tablet and recognizing voice information to be spoken,
Only significant words can be extracted and attached as medium information, real-time processing by simple operation is possible, and operation skill is not required. Furthermore, by the customer operating in accordance with a predetermined procedure, processing by the customer alone is possible.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. First, the first embodiment will be described. FIG. 2 shows an example of the commuter pass purchase application P according to the first embodiment. Name entry field 1, gender / age entry field 2, address entry field 3, work entry field 4, boarding section A (use section) entry field 5, a new / continuation classification entry field 6 and a usage start date / effective period entry field 7 are provided.

FIG. 1 schematically shows a configuration of a medium issuing device for printing and issuing, for example, a commuter pass according to the first embodiment. In FIG. 1, reference numeral 11 denotes a microphone (hereinafter simply referred to as a microphone) as a voice input unit,
Enter the voice information generated by the staff at the station. Microphone 1
The voice information input from 1 is sent to the voice recognition unit 12 as a voice recognition means. The voice recognition unit 12 recognizes the voice information input using the syllable dictionary 13 in syllable units.

The recognition result of the voice recognition section 12 is sent to the word recognition section 14 as word recognition means. Word recognition unit 14
Performs word recognition processing on the recognition result of the voice recognition unit 12 using the word dictionary 15. The recognition result of the word recognition unit 14 is sent to the main control unit 16 mainly composed of a CPU and the like.

The main control unit 16 controls the entire apparatus, and includes a keyboard 17, a CRT display unit 18, an image input scanner 19 as an image input unit, and an issuing unit 20 as an issuing unit for issuing a medium. Etc. are connected respectively. The image input scanner 19 receives the application form P shown in FIG. 2, optically reads the image, and inputs the image. The issuing unit 20 prints and issues a commuter pass (medium) C based on various input information.

The overall operation of such a configuration will be described briefly. The clerk receives the application form P in which the customer has filled in the necessary items, inserts it into the image input scanner 19 and sets it, and inputs the image of the application form P. After that, the clerk, while visually checking the contents of the application form P, asks the customer, for example, in the case of the application form P in FIG. 2, "From Kawasaki station to Shibuya station", "Shinagawa via", "It's been 6 months since October 1st. "

As described above, when the staff interactively confirms with the customer, the voice information uttered by the staff is input from the microphone 11 and sent to the voice recognition unit 12. The voice recognition unit 12 separates the voice information input from the microphone 11 into syllable units,
The syllable dictionary 13 is used to perform recognition processing in syllable units. The recognition result for each syllable unit of the voice recognition unit 12 is sent to the word recognition unit 14. The word recognition unit 14 calculates the similarity between the recognition result for each syllable unit and the dictionary data of the word dictionary 15, and sends the result to the main control unit 16.

The main control unit 16 displays the recognition result from the word recognition unit 14 on the display unit 18. The staff member confirms the recognition result displayed on the display unit 18, and after confirming the recognition result, presses the issuance start key (not shown) in the keyboard 17 to cause the main control unit 16 to display the input information. Based on this, the issuing unit 20 is operated to print and issue the commuter pass C as shown in FIG. 3, for example, and the issued commuter pass C is provided to the user.

FIG. 4 is a block diagram of the word recognition unit 14 in FIG. 1, which will be described in detail below with reference to this figure.
The recognition result of the voice recognition unit 12 is stored in the recognition result buffer 31 as recognition enumeration information (hereinafter sometimes simply referred to as recognition information). Under the control of the first control unit 32, the recognition enumeration information in the recognition result buffer 31 is divided into word units and transferred to the keyword similarity calculation unit 33. The keyword similarity calculation unit 33 makes correspondence according to the order of the application form P to be uttered.

[0023] Specifically, for example, first, the staff is "Kawasaki Karashibuyaeki Madedesne" (Kawasaki Station to Shibuya Station) or "Kawasaki Karashibuyamade Desne" (Kawasaki to Shibuya). Say). The keyword similarity calculation unit 33 uses the keyword dictionary 34 to perform word recognition on the recognition enumeration information in the recognition result buffer 31 according to the utterance order of the staff.

First, a keyword search (word spotting) for "exhaust" or "empty" is performed. As a result, when one of the keywords is determined, a significant station name word is recognized by recognizing the previous recognition information using the unique word dictionary 35.

That is, the keyword dictionary 34 stores, for example, words as shown in FIG. 5 in advance. Therefore, the first control unit 32 causes the words “exch” “a1” and “ch” “a2” stored in the keyword dictionary 34.
And the word string created from the recognition information in the recognition result buffer 31 is calculated by the keyword similarity calculation unit 33. As a result, m5, m6, m of the recognition result buffer 31
Detects 7 or m8-bit "Exkara" as a keyword.

The first control unit 32 recognizes the keyword "Exkara" to determine that there is a station name in front of it, and the second control unit 36 informs the position information (m5, m6, m7) of the station name.
, M8). The second control unit 36 is "exciting"
, "M", "m", "m2", "m3", and "m4" of "Kawasaki" before are transferred to the unique word similarity calculator 37,
Similarity calculation is performed between and. As a result, "Kawasaki"
Is recognized as a station name word and is transferred to the second control unit 36.

Similarly, when the keyword "ekimade" is recognized, the unique word of the station name is recognized and "shibuya" is recognized.
Is recognized as a station name word and is transferred to the second control unit 36. The unique word dictionary 35 stores, for example, station names and numerical information, and switches between groups A and B in accordance with the result of keyword search by the keyword similarity calculation unit 33. For example, group A stores station names, and group B stores numerical information such as year, month, and day.

Next, when the staff member says "Shinagawa Keiyu Desne (via Shinagawa)", "Keiyu" is searched by keyword, and "Shinagawa" in the recognition result buffer 31 is calculated in the unique word similarity calculation unit 37. Recognize words,
The significant information is transferred to the second controller 36.

Furthermore, when the staff member says, "Gyutsutsu Itachikara Rokka Getsudesune (6 months from October 1st)," by searching for "Gatsu", "Tachi", and "Getsu" respectively. , "Ju", "Tsui",
The "rocker" is recognized as a unique word and transferred to the second control unit 36 as significant information.

FIG. 6 is a flow chart for explaining the word processing described above. By using "Desune" as the ending word of the phrase, the next phrase information can be predicted. Also, if the individual keyword is uncertain, "color, made",
By using the utterance order of multiple words such as "Nen, Gatsu, Nichi" as a knowledge dictionary and correcting uncertain words, the unique words can be surely recognized. The unique words in this example are "Kawasaki", "Shibuya", "Shinagawa", "10",
They are "1" and "6".

Then, the main control section 16 finally recognizes the images of the name entry field 1 and the sex / age entry field 2 cut out from the image of the application P input from the image input scanner 19 as the recognized unique word. And is issued by the issuing unit 20 as the commuter pass C shown in FIG.

As described above, according to the above-described first embodiment, a natural dialogue voice to the customer of the clerk is input and recognized in response to the customer's request in the window reception work. Real-time processing is possible,
It does not require skill in operation as in the past. Furthermore, since the specific word of the specific speaker of the staff is the recognition target, the efficiency of information input can be improved.

Next, a second embodiment will be described. FIG. 7 schematically shows the configuration of a medium issuing device for printing and issuing, for example, a commuter pass according to the second embodiment. Note that the same parts as those in FIG. In FIG. 7, 11a is a microphone as a voice input means for a staff member, which inputs voice information generated by a staff member at a station counter. Reference numeral 11b is a microphone as a voice input means for the customer, and inputs voice information generated by the customer who visits the station counter. The voice information input from the microphones 11a and 11b is
Each is sent to the voice input selection unit 21. The voice input selection unit 21 normally outputs both voice information input from both microphones 11a and 11b as both inputs are in the pass state. However, by operating the selection switch 22, one of the microphones It is designed to switch to voice input only.

Each voice information output from the voice input selection unit 21 is sent to the voice recognition unit 12 as a voice recognition means. The voice recognition unit 12 recognizes the voice information input using the syllable dictionary 13 in syllable units.

The recognition result of the voice recognition section 12 is sent to the word recognition section 14 as word recognition means. Word recognition unit 14
Performs word recognition processing on the recognition result of the voice recognition unit 12 using the word dictionary 15. The recognition result of the word recognition unit 14 is sent to the main control unit 16 mainly composed of a CPU and the like.

The main control unit 16 controls the entire apparatus, and has a keyboard 17, a CRT display unit 18, an issuing unit 20 as an issuing unit for issuing a medium, and a manual input for manually inputting character information and the like. A display-integrated tablet 23 as a means is connected to each. Issuer 2
0 is to print and issue a commuter pass (medium) based on various input information.

The display-integrated type tablet 23 is, for example,
It is operated by the customer at the station window, and is configured as shown in FIG. That is, on the surface of the tablet 23, the entry items (name entry area 8, gender entry area 9, age entry area 10) as shown in FIG. 8 are preprinted, and the customer uses a specific pen 24 provided. And fill in the prescribed positions. The written character image is on the tablet 2
3, the images of the name entry area 8, the gender entry area 9, and the age entry area 10 are cut out, so that the main control unit 16
To be forwarded to.

The overall operation of such a structure will be briefly described. FIG. 8 shows an example of items described when purchasing a commuter pass. The clerk requests the customer to fill in predetermined items on the display-integrated tablet 23. The customer uses the pen 24 to enter the name and age, and marks either the man or the woman. The character image written on the display-integrated tablet 23 is the main control unit 16
Is forwarded to

Next, the staff member responds to the customer in a question form. For example, in response to the staff member "Where to where?", The customer replies, "From Kawasaki to Shibuya," which section is on board. Also, answer "Yes" to "Shinagawa via," and "6 months from October 1st" to "When has it been some months?" The microphone 11a inputs voice information uttered by a staff member, and the microphone 11b inputs voice information uttered by a customer.

Voice information from the microphone 11a and the microphone 11
The voice information from b is normally input to the voice recognition unit 12 via the voice input selection unit 21 which is in the pass state. Here, as an example, description will be made assuming that the clerk and the customer alternately speak in a question-and-answer form.

First, the voice information of the clerk input from the microphone 11a is input to the voice recognition unit 12 via the voice input selection unit 21. The voice recognition unit 12 separates the input voice information into syllable units, and uses the syllable dictionary 13 to perform recognition processing in syllable units. The recognition result for each syllable unit from the voice recognition unit 12 is sent to the word recognition unit 14. The word recognition unit 14 recognizes the recognition result for each syllable unit and the word dictionary 15.
Of the dictionary data is calculated, and the result is sent to the main control unit 16.

Next, the voice information of the customer input from the microphone 11b is input to the voice recognition unit 1 via the voice input selection unit 21.
2, the same processing as above is performed, and the main control unit 16
Sent to The main control unit 16 displays the recognition result from the word recognition unit 14 on the display unit 18. Then, when the input of all the voice information is completed, the staff member confirms the recognition result displayed on the display unit 18, and after confirming the result, presses the issuing start key (not shown) in the keyboard 17, The main control unit 16 determines the issuing unit 2 based on the input information.
0 to operate, for example, a commuter pass C as shown in FIG.
Is printed and issued, and the issued commuter pass C is provided to the user.

FIG. 10 is a block diagram of the word recognition unit 14 in FIG. 7, which will be described in detail below with reference to this figure. The recognition result of the voice recognition unit 12 is stored in the recognition result buffer 31 as recognition enumeration information (hereinafter sometimes simply referred to as recognition information). Under the control of the first control unit 32, the recognition enumeration information in the recognition result buffer 31 is divided into word units and transferred to the keyword similarity calculation unit 33.

Specifically, for example, the staff member first says "Doco Calado Comadedesca". The keyword similarity calculation unit 33 performs word recognition by calculating the similarity between a predetermined word string in the recognition result buffer 31 and the dictionary data of the keyword dictionary 34 according to the utterance order of the staff. Here, in the keyword dictionary 34, for example,
Words as shown in FIG. 11 are stored in advance.

First, “docokora” and “docomade” in the keyword dictionary 34 are searched for keywords (word spotting) in the keyword similarity calculation section 33.
The first control unit 32 receives the result of the keyword similarity calculation unit 33 and transfers the information to the second control unit 36. The second control unit 36 interprets the meanings of the keywords “dococara” and “docomade” and predicts that the customer will utter the station name next.

Then, in response to the answer "Kawasaki Karashibu Yamadedes" uttered by the customer, the keyword similarity calculation unit 33 calculates similarity between the recognition information in the recognition result buffer 31 and the dictionary data of the keyword dictionary 34. First,
Search for "Kara" as a keyword. First control unit 32
Transfers the "color" information to the second control unit 36, so that the second control unit 36 determines that there is a station name in front of it, and the unique word similarity calculation unit 37 causes the recognition result buffer 31
Then, the similarity between the recognition information in the dictionary and the dictionary data of the unique word dictionary 35 is calculated. As a result, “Kawasaki” is recognized as the station name word and is transferred to the second control unit 36.

After that, similar processing is applied to the continuing voice information. That is, the first control unit 32 uses the word “dococara” a stored in the keyword dictionary 34 as a keyword, and searches (similarity calculation) while creating a word string to determine whether the word is stored in the recognition result buffer 31. As a result, m1, m2, m3, m of the recognition result buffer 31
Detect 4-bit "dococara" as a keyword.

Furthermore, the keyword similarity calculator 33
The similarity is calculated between the word “docomade” b in the keyword dictionary 34 and the word string in the recognition result buffer 31, and “docomade” is detected as a word.

Next, the first control unit 32 responds to the customer "Kawasaki Karashibu Yamadedes" by the recognition result buffer 31.
By detecting the n5 and n6 bit "color" keyword, it is determined that there is a station name in front of it, and the position information (n5, n6) is transmitted to the second control unit 36. The second control unit 36 transfers n1, n2, n3, and n4 bits of “Kawasaki” before “kara” to the unique word similarity calculation unit 37, and performs similarity calculation with the unique word dictionary 35. . As a result, "Kawasaki" was recognized as the station name word, and the second
It is transferred to the control unit 36.

Similarly, when the keyword "Made" is recognized, the unique word of the station name is recognized, "Shibuya" is recognized as the station name word, and the word is transferred to the second control unit 36.
The unique word dictionary 35 stores, for example, station names and numerical information, and is automatically stored in the dictionary when the customer's unique word similarity is calculated according to the result of the keyword search by the staff member's keyword similarity calculation unit 33. The A group and the B group are switched to. For example, group A stores station names, and group B stores numerical information such as year, month, and day.

Next, when the staff member says "Shinagawa Keiyu Desne (via Shinagawa)", "Keiyu" is searched for by keyword, and "Shinagawa" in the recognition result buffer 31 is calculated in the unique word similarity calculation unit 37. Recognize words,
The significant information is transferred to the second controller 36.

Furthermore, when the staff member answers the question "Itsukaranan Kagetsudeska", the customer answers "Juugatsutsu Itachikara Rakkagetsude (6 months from October 1st)". By performing a keyword search for “Tachi” and “Gets” respectively, the unique words of “Ju”, “Tsui” and “Rocca” are recognized, and the second control unit 3 recognizes them as significant information.
Transfer to 6.

FIG. 12 is a flow chart for explaining the word processing described above. By using "Desune" as the ending word of the phrase, the next phrase information can be predicted. In addition, FIG. 13 is a flowchart illustrating an example of processing procedures of the above-mentioned staff and customer.

Finally, "Kawasaki", "Shibuya", "Shinagawa", "10", "1", and "6" are transferred to the main control unit 16 as significant word information. Then, the main controller 16
The recognized word and the character image of the name, gender and age input from the tablet 23 are combined and issued by the issuing unit 20 as the commuter pass C shown in FIG.

As described above, according to the above-described second embodiment, when issuing a commuter pass or a personal card in the window reception work, interactive information exchange between the staff and the customer, that is, character information by the tablet is performed. input,
By recognizing the spoken information to be spoken, only significant words can be extracted and attached as medium information, real-time processing by simple operation is possible, and operation skill required in the past is not required. Moreover, the application for purchasing a commuter pass, which was required in the past, is no longer necessary.

Next, a third embodiment will be described. FIG. 14 schematically shows the configuration of a medium issuing device for printing and issuing, for example, a commuter pass according to the third embodiment. The same parts as those in FIG. 7 are designated by the same reference numerals, and the description thereof will be omitted. Only different parts will be described in detail. The third embodiment is the same as the second embodiment shown in FIG.
An online character recognition unit 25 for recognizing character information input from the tablet 23 is added between the display-integrated tablet 23 and the main control unit 16.

In this case, an entry frame (name entry frame 41, gender entry frame 42, age entry frame 43) as shown in FIG. 15 is preprinted on the tablet 23, and the customer enters the name entry frame. Name in 41, age entry frame 43
The age is entered therein, and a mark is put in the gender entry frame 42 for confirmation. The character recognition unit 25 performs character recognition using a dictionary of kanji and kana in the name entry frame 41 from the position information given in advance, and the age entry frame 4
3 is set to recognize characters by using a number dictionary.

In addition, the character recognition unit 25 detects which of the men and women in the gender entry frame 42 is marked by the image information in the gender entry frame 42, and the gender set in the main control unit 16 is detected. It is designed to put a circle on either of the.

The recognition result of the character recognition unit 25 is the result of the main control unit 1.
6 is transferred. The main control unit 16 issues a commuter pass C as shown in FIG. 16 by combining and editing the recognition result of the character recognition unit 25, the circle mark for gender discrimination, and the above-mentioned voice recognition result. It is issued by the section 20.

In the description of the second and third embodiments, both the microphones 11a and 11b are in the passing state when the voice input selection section 21 is normally used. It is possible to switch between voice input from only the microphone 11a and voice input from only the microphone 11b. In this case, by switching to only the microphone 11b, the customer can freely operate the device even if a staff member is absent.

That is, the input from the tablet 23 and the voice input from the microphone 11b are input, for example, as shown in FIG.
7, the second control unit 36 analyzes the keyword calculation and unique word calculation described in FIG. 10 and transfers them to the main control unit 16. Then, the commuter pass C shown in FIG. 9 is issued.

According to this, by the customer operating in accordance with a predetermined procedure, the processing by the customer alone can be performed. Further, in the above-described embodiment, the case where the commuter pass is printed and issued has been described, but the present invention is not limited to this, and for example, a normal ticket (e.g., a ticket or an admission ticket) or a bank The same applies to the case of issuing a bankbook or credit card used.

[0063]

As described above in detail, according to the present invention, it is possible to provide a medium issuing device which enables real-time processing by a simple operation and does not require any skill in operation.

[Brief description of drawings]

FIG. 1 is a block diagram schematically showing a configuration of a medium issuing device that prints and issues a commuter pass according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an example of a commuter pass purchase application form.

FIG. 3 is a diagram showing an example of a issued commuter pass.

FIG. 4 is a block diagram schematically showing a configuration of a word recognition unit in FIG.

FIG. 5 is a diagram showing an example of keyword words stored in a keyword dictionary.

FIG. 6 is a flowchart illustrating word processing.

FIG. 7 is a block diagram schematically showing the configuration of a medium issuing device for printing and issuing a commuter pass according to a second embodiment of the present invention.

FIG. 8 is a diagram illustrating an example of entry items on a tablet.

FIG. 9 is a diagram showing an example of an issued commuter pass.

10 is a block diagram schematically showing a configuration of a word recognition unit in FIG.

FIG. 11 is a diagram showing an example of keyword words stored in a keyword dictionary.

FIG. 12 is a flowchart illustrating word processing.

FIG. 13 is a flowchart illustrating an example of a processing procedure of an employee and a customer.

FIG. 14 is a block diagram schematically showing the configuration of a medium issuing device that prints and issues a commuter pass according to a third embodiment of the present invention.

FIG. 15 is a diagram illustrating an example of filling in a tablet for online character recognition.

FIG. 16 is a diagram showing an example of a commuter pass issued in consideration of online character recognition.

FIG. 17 is a flowchart illustrating an example of a procedure of customer-only processing.

[Explanation of symbols]

P: Commuter ticket purchase application, 11, 11a, 11b ... Microphone (voice input means), 12 ... Voice recognition unit (voice recognition means), 13 ... Syllable dictionary, 14 ... Word recognition unit (word recognition) Means), 15 ... word dictionary, 16 ... main control unit, 1
9 ... Image input scanner (image input means), 20 ... Issuing unit, C ... Commuter pass (medium), 21 ... Voice input selecting unit, 22 ... Selection switch, 23 ... Display integrated tablet (manual) Input means), 25 ... Online character recognition section (character recognition means), 31 ... Recognition result buffer, 32 ...
... First control unit, 33 ... Keyword similarity calculation unit, 34
…… Keyword dictionary, 35 …… Unique word dictionary, 36 ……
Second control unit 37 ... Unique word similarity calculation unit.

Claims (9)

[Claims] 1. The medium issuance condition information is received by a staff member who receives the application form in which the medium issuance condition information is entered, which is presented by the customer, and by inputting the medium issuance condition information described in the received application form. A medium issuing device for issuing a medium according to, when a staff member who received the application form interactively confirms with the customer the meaningful clauses in the medium issuing condition information described in the application form, the staff member Voice input means for inputting voice information to be uttered, voice processing means for detecting a significant word by recognizing the voice information input by the voice input means, and a predetermined medium based on the processing result of the voice processing means. A medium issuing device comprising: an issuing unit for issuing the. 2. The medium issuance condition information is obtained by the staff member receiving the application form in which the medium issuance condition information is entered, which is presented by the customer, and by inputting the medium issuance condition information described in the received application form. A medium issuing device for issuing a medium according to, when a staff member who received the application form interactively confirms with the customer the meaningful clauses in the medium issuing condition information described in the application form, the staff member Voice input means for inputting voice information to be uttered, voice recognition means for recognizing the voice information input by this voice input means, and specific part-of-speech or phrase following the significant word for the recognition result of this voice recognition means. A medium issuing device comprising: a word recognizing means for recognizing a significant word by searching; and an issuing means for issuing a predetermined medium based on a recognition result of the word recognizing means. 3. The word recognizing means, when the searched specific part of speech or phrase is unclear, corrects the relationship with another specific part of speech or phrase within one phrase. Media issuing device. 4. A manual input means for a customer to manually input a part of the information displayed on the medium, and a question and answer performed between the staff and the customer based on the information input by the manual input means. A voice input means for inputting voice information, a voice processing means for detecting a significant word from a semantic analysis of utterance by recognizing the voice information input by the voice input means, a processing result of the voice processing means and the manual operation. A medium issuing device comprising: an issuing unit that issues a predetermined medium based on a combination of information input by the input unit. 5. A manual input means by which a customer manually inputs a part of the information displayed on the medium, and a question and answer performed between the clerk and the customer based on the information input by the manual input means. A voice input means for inputting voice information, a voice recognition means for recognizing the voice information input by the voice input means, and a recognition result of the voice recognition means for predicting a customer's answer to a question of a staff member, A word recognition means for recognizing a significant word by searching for a specific part of speech or a phrase following the word, and a predetermined medium is issued based on the combination of the recognition result of this word recognition means and the information input by the manual input means. A medium issuing device, comprising: 6. A tablet-type manual input means for a customer to manually input a part of the character information displayed on a medium, a character recognition means for recognizing the character information input by the manual input means, and the manual operation. Based on the character information input by the input means, a voice input means for inputting voice information of a question and answer performed between the clerk and the customer; and a voice recognition by recognizing the voice information input by the voice input means. A voice processing means for detecting a significant word from the semantic analysis; and an issuing means for issuing a predetermined medium based on a combination of the processing result of the voice processing means and the recognition result of the character recognition means. Issuing device. 7. A tablet-type manual input means for a customer to manually input a part of the character information displayed on a medium, a character recognition means for recognizing the character information input by the manual input means, and the manual operation. A voice input means for inputting voice information of a question and answer performed between the staff and the customer based on the character information input by the input means, and a voice recognition means for recognizing the voice information input by the voice input means. The word recognition means for recognizing the significant word by predicting the customer's answer to the staff's question with respect to the recognition result of the voice recognition means and searching for a specific part of speech or phrase following the significant word, and the word recognition means. A medium issuing device for issuing a predetermined medium based on a combination of the recognition result of 1. and the recognition result of the character recognizing device. 8. A manual input means for a customer to manually input a part of information displayed on a medium, a voice input means for the customer to input voice information according to a predetermined procedure, and the voice input means. A voice processing means for detecting a significant word by recognizing the voice information input by, and issuing a predetermined medium based on a combination of the processing result of the voice processing means and the information input by the manual input means. A medium issuing device comprising: 9. A tablet-type manual input means for a customer to manually input a part of the character information displayed on a medium, a character recognition means for recognizing the character information input by the manual input means, and the customer. Is a voice input means for inputting voice information according to a predetermined procedure, a voice processing means for detecting a significant word by recognizing the voice information input by the voice input means, and a processing result of the voice processing means. And a issuing unit for issuing a predetermined medium based on a combination of the recognition results of the character recognizing unit, and a medium issuing device.