JP5079893B2 - Data input device and method, and program - Google Patents

Description

  The present invention relates to a data input device using voice recognition technology, a data input method for a computer having voice recognition means, and a program for causing a computer having voice recognition means to function as the data input device of the invention.

  In recent years, a speech recognition technique using speech instead of a keyboard or the like as an input means has been developed and is being put into practical use in various fields. Speech recognition is a mechanism in which a speech pattern that is a feature amount of speech is calculated from an input speech signal, and a speech pattern having a high degree of similarity is extracted after comparison with a standard pattern prepared in advance.

A data input device that uses such voice recognition technology to process data input according to customer orders in restaurants such as restaurants is already known (see, for example, Patent Document 1). .
JP 2004-326513 A

  In general, however, current speech recognition technology has a high probability of not being recognized unless it is spoken in the same way as a standard pattern speech. That is, the recognition rate is easily influenced by the content of human speech. On the other hand, there are many menu products with similar item names such as “ramen”, “tanmen”, “tantanmen”, etc. in restaurants. In order to reliably recognize the voices of these item names, it is necessary to speak accurately and carefully one sound at a time. For this reason, there is a concern that the voice input of the order takes time. In addition, in restaurants where customers of various customers come and go, there are cases where voices cannot be recognized due to the influence of surrounding noise and conversation. Moreover, if you utter an item name that includes a sound or sound repellent, such as “Be” of “Beijing Duck” or “N” or “tsu” of “Hamburg Set”, the sound will be temporarily silent, that is, no sound, In some cases, it may be recognized that the voice has ended before “n” or “tsu” and erroneous recognition occurs. From this point of view, it has been considered that applying speech recognition technology to an order input device for restaurants with customer service is not practical.

  The present invention has been made on the basis of such circumstances, and the object of the present invention is not easily affected by ambient noise and conversations, and many of them are similar to the content of utterances, and include sounding and repelling. Therefore, it is an object of the present invention to provide a data input device, a method thereof, and a program which can input data reliably using voice even when there are many cases, and can simplify data input work.

The data input device according to the present invention includes an audio menu file for storing an item name for identifying a plurality of items to be ordered, an item code set corresponding to the item name, and audio data of the item name, and an audio Voice input means for inputting a voice, an utterance key that is operated continuously while voice is input by the voice input means, and an input section detection that detects that the voice key is continuously operated as a voice input section means, a speech recognition means for recognizing the speech pattern input through the voice input means to said speech input in the interval detected by the input section detecting means, a sound pattern data recognized by said speech recognition means the sought item name voice data and match rates compared individually audio menu each ordered to products stored in a file, the matching ratio is preset threshold A similar name search means for extracting a plurality of item-name sound data of more than, similar to get the goods codes and item names corresponding to a plurality of item name audio data extracted by said similar name retrieval means from said voice menu file a data acquisition means, a candidate menu display means for said selectably list item name of the each candidate for a plurality of item name candidates obtained by the similar data acquisition unit sequentially higher the match rate, the candidate menu display means When an arbitrary item name is selected from the plurality of item names displayed by the above, voice input data for processing the item code read from the voice menu file based on the selected item name as order input data Processing means and an order name including the item name and the order score processed as the order input data by the voice input data processing means. Order menu display means for displaying a list of order input data as an order menu, a confirm key for confirming the order input data displayed in a list on the order menu display means, and the order menu display when the confirm key is operated Transmission means for transmitting order input data including an item code and an order score for the item name displayed by the means to a station that stores and manages the order data.

Further, the data input method according to the present invention includes a voice menu file for storing audio data item name to identify a plurality of orders goods for each and the item name goods code set in correspondence with the item name comprises an audio input means for inputting a voice, while inputting a speech by the speech input means, a speech key to be operated continuously, and speech recognition means for recognizing a voice input through the voice input means A computer data input method, wherein the computer detects a voice input section while the utterance key is continuously operated, and recognizes a voice pattern input in the detected voice input section. and obtains the recognized item name voice data and match rates compared individually for each ordered to goods stored in the voice menu file in the voice pattern data, the one Rate extracts a plurality of item name audio data exceeding the preset threshold value, respectively on the extracted plurality of line items audio data to obtain a corresponding article code and item name from said voice menu file, obtained and a plurality of item name candidate item name of each of the candidate and selectably listed in order the matching rate is high, when an arbitrary item name from among a plurality of line items that are the list is selected, the The item code read from the voice menu file based on the selected item name is processed as order input data, a list of the order input data including the item name and the order score is displayed as an order menu, and the enter key is operated. The order input data including the item code and the order score for the item name displayed on the order menu display means is stored and managed. And transmits to the station that.

In addition, the program according to the present invention includes an audio menu file for storing an item name for identifying a plurality of items to be ordered, an item code set corresponding to the item name , and audio data of the item name , A voice input section is a period during which the utterance key is continuously operated in a computer provided with a voice input means for inputting the voice and an utterance key that is operated continuously while the voice is input by the voice input means. An input section detection function to be detected , a voice recognition function for recognizing a voice pattern input via the voice input means within the voice input section detected by the input section detection means, and a voice recognition means sought item name voice data and match rates compared individually for each ordered to goods stored in the voice menu file in the voice pattern data, said matching There the similarity name search function for extracting a plurality of item names audio data exceeding the preset threshold value, the goods codes and item names corresponding to a plurality of item name audio data extracted by said similar name search function Similar data acquisition function acquired from the voice menu file, and candidate menu display for selecting a plurality of item name candidates obtained by the similar data acquisition function so that the item names of each candidate can be selected in descending order of the match rate order function and, when an arbitrary item name from among a plurality of item name displayed by the candidate menu display function is selected, the article code read from said voice menu file based on the selected item name and the Includes voice input data processing function to be processed as input data, and item name and order score processed by the voice input data processing function And order menu display function of displaying a list of over Da input data as the order menu, when the enter key to confirm the order entry data listed in the order menu display function is operated is displayed in the order menu display function This is to realize a transmission function for transmitting the order input data including the item code and the order score for the item name to the station that stores and manages the order data .

  According to the present invention in which such measures are taken, it is difficult to be affected by ambient noise, conversation, etc., and voice is used even when there are many things similar to the content of the utterance or often contain prompting sounds and sound repellents. As a result, data can be input reliably, and the data input operation can be simplified.

The best mode for carrying out the present invention will be described below with reference to the drawings.
This embodiment is a case where the present invention is applied to a data input device that processes order data input of a menu item ordered by a customer at a restaurant, so-called order input device.

  FIG. 1 is a perspective view showing an external appearance of an order input device 1 according to the present embodiment, and FIG. 2 is a block diagram showing a main configuration of the order input device 1. The order input device 1 includes a device main body 11 and a display unit 12. A microphone 13 is provided as a voice input means for inputting voice. The display unit 12 is a touch panel type composed of a display 121 using a liquid crystal and a touch panel sensor 122, for example. Note that the microphone 13 is not limited to the one protruding from the apparatus main body 11 as shown in the figure, and may be built in the display unit 12, for example.

  The apparatus main body 11 includes a CPU (Central Processing Unit) 111 that plays a central part of the computer, a ROM (Read Only Memory) 112 in which fixed data such as a program for controlling the operation of the CPU 111 is stored in advance, and data by the CPU 111. RAM (Random Access Memory) 113 to which data is written and read, HDD (Hard Disk Drive) 114 as a large-capacity storage medium, and electronic devices connected via a communication network such as a LAN (Local Area Network) A communication interface 115 that controls data communication, a display controller 116 that controls screen display of the display 121, a touch panel interface 117 that receives a touch detection signal from the touch panel sensor 122, and an analog audio signal that is input from the microphone 13. Digital signal Speech recognition engine 119 to perform speech recognition based on the digital audio signal output from the A / D converter 118 and the A / D converter 118 for converting is mounted on. The CPU 111, the ROM 112, the RAM 113, the HDD 114, the communication interface 115, the display controller 116, the touch panel interface 117, and the voice recognition engine 119 are electrically connected by a bus line BL such as an address bus or a data bus.

  The communication interface 115 is connected to a station for storing and managing each customer's order data. Also, electronic devices such as a kitchen terminal that gives instructions for cooking to the kitchen and a POS terminal that processes accounting for food and drink are connected.

  As shown in FIG. 3, the HDD 114 stores various data files such as a menu master file 31, an audio menu file 32, and an acoustic dictionary file 33.

  The menu master file 31 corresponds to each menu item code for identifying each of the various menu items provided at the restaurant where the order input device 1 is used, and is identified by the menu item code. Various data relating to the menu product such as the item name and unit price of the menu product are stored.

  In the voice menu file 32, a standardized value is obtained by quantifying the item name of the menu item specified by the menu item code and the voice feature amount when the item name is spoken in correspondence with each menu item code. Item name voice data which is a simple voice pattern is stored. Here, the voice menu file 32 includes a target name (item name) for identifying a plurality of targets (order target products), a target code (product code) set corresponding to the target name, and a voice of the target name. Storage means for storing data is configured. The item name data in the menu master file 31 may be used without storing the item name data in the voice menu file 32.

  The acoustic dictionary file 33 stores phrase audio data, which is standard audio pattern data obtained by quantifying the voice feature amount when a predetermined word is uttered. The difference between the voice menu file 32 and the acoustic dictionary file 33 will be described. The acoustic dictionary file 33 is a data file used when the voice recognition engine 119 recognizes the voice input from the microphone 13. The voice menu file 32 is a data file used when the CPU 111 extracts item name voice data similar to the voice pattern data recognized by the voice recognition engine 119.

  For example, in the case of three items of menu products “ramen”, “tanmen”, and “tantanmen”, the audio dictionary file 33 stores the phrase voice data of the phrase “ra”, the phrase “men”, and the phrase “tan”. ing. On the other hand, the voice menu file 32 stores item name voice data obtained by combining two phrase voice data of the phrase “Ra” and the phrase “Men” for the menu code of the menu item “Ramen”. For the menu code of the menu product "Tanmen", the item name voice data that combines the two phrases audio data of the phrase "Tan" and the phrase "Men" is stored. On the other hand, item name voice data obtained by combining three phrase voice data of the phrase “tan”, the phrase “tan”, and the phrase “men” is stored.

  In the order input device having such a configuration, a program for realizing a voice order data input function is stored in the ROM 112. As memory areas necessary for executing this program, as shown in FIG. 4, each of the memory areas of the input voice data memory 41, the comparison voice data memory 42, and the order candidate table 43 is formed in the RAM 113. Yes. Incidentally, the order candidate table 43 can write menu item code, item name, and match rate data corresponding to the table numbers set in ascending order from 1. Note that the program may be stored in the HDD 114 and read out by the CPU 111 at the time of activation and expanded in the program area of the RAM 113.

  Thus, for example, when the power is turned on, the order input device 1 starts up, and when the order input job is selected from the job menu, the program is started. Thus, the CPU 111 processes customer order data input according to the procedure shown in the flowchart of FIG.

  First, the CPU 111 causes the display 121 to display an initial information input screen regarding order input such as seat information, number of persons, person-in-charge information as ST (step) 1. Further, the order memory (not shown) formed in the RAM 113 is cleared. Next, the CPU 111 waits for the input of the initial information as ST2. If it is detected by the signal from the touch panel sensor 122 that the seat information, the number of persons, the person in charge information, and the like are input, the CPU 111 switches the screen of the display 121 to the order input screen 50 as ST3.

  The items of initial information are not limited to the three items of seat information, number of persons, and person-in-charge information, and other items may be combined. Any item may be omitted. Alternatively, the initial information may be omitted, and the initial screen after rising may be the order input screen 50.

  An example of the order input screen 50 is shown in FIG. As shown in the figure, the order input screen 50 includes a plurality of menu classification keys 52, a plurality of menu keys 53, a multiplication key 54, a confirmation key 55, a cancel key 56, a voice in addition to the numeric keys 51 of “0” to “9”. Touch key images such as the recognition key 57 are displayed. Further, an order menu column 58 for confirming the order input content, a number display column 59 for displaying the number data, and the like are formed. Each menu classification key 52 is assigned an item for classifying various food menu items such as a single item menu, a set menu, a beverage menu, and a dessert menu. Each menu key 53 is assigned an item of various food menu items belonging to the menu category selected by the touch operation of the menu category key 52.

  After displaying the order input screen 50, the CPU 111 determines whether or not voice order data input is declared in ST4. If the order data input by voice is not declared, the CPU 111 determines whether or not the order data input by key operation is performed at ST5. If no order data is input by key operation, the CPU 111 proceeds to the process of ST7.

  If it is detected in ST5 that any of the menu keys 53 is touched by a signal from the touch panel sensor 122, the menu item assigned to the touched menu key is selected. Since order data input has been declared, the CPU 111 executes an order data input process as ST6. That is, the menu item code and the order number n of the menu item are registered in the order memory (not shown) of the RAM 113. Further, the item name and the order score n of the menu product are displayed in the order menu column 58. The order number n is set to “1” as an initial value, and when a multiplier is input by the ten key 51 and the multiplication key 54, the order number is the multiplier. Thereafter, the CPU 111 proceeds to the process of ST7. In ST7, the CPU 111 determines whether confirmation of order data input has been declared. If confirmation of order data input is not declared, the process returns to ST4.

  If it is detected in ST7 that the confirmation key 55 has been touched by a signal from the touch panel sensor 122, confirmation of order data input has been declared, so the CPU 111 determines whether or not the order data has been input in ST8. to decide. If no data is registered in the order memory, the order data has not been input, and the process returns to ST4.

  If data has been registered in the order memory in ST8, since the order data has already been input, the CPU 111 switches the screen of the display 121 to the order confirmation screen in ST9. Although not shown, on the order confirmation screen, the item name, the order score, the price, the total amount, and the like of the menu product for which order data has been input are displayed. In addition, a touch key image of a confirmation key that declares confirmation of the order is displayed.

  Here, if it is detected in ST10 that the confirmation key has been touched by a signal from the touch panel sensor 122, the CPU 111 registers the station connected through the communication interface 115 in ST11 in the order memory. Order data is sent. This completes the order input process for one customer, and the CPU 111 returns the screen of the display 121 to the initial information input screen.

  When it is detected in ST4 that the voice recognition key 57 has been touched by a signal from the touch panel sensor 122, since order data input by voice is declared, the CPU 111 displays the screen of the display 121 as a voice order in ST12. Switch to the input screen 60.

  An example of the voice order input screen 60 is shown in FIG. As shown in the figure, on the voice order input screen 60, touch key images such as a plurality of candidate menu keys 62, a multiplication key 63, a confirmation key 64, a cancel key 65, a voice recognition end key 66, a speech key 67, etc. in addition to the numeric keypad 61. Is displayed. Also, an order menu column 68 for confirming the order input content, a number display column 69 for displaying the number data, and the like are formed. In the plurality of candidate menu keys 62, no item of the menu product is displayed at this time. In the utterance key 67, a message “Please speak while pressing this button” is displayed. Here, the CPU 111 constitutes screen control means for displaying the utterance key 67, that is, the touch key image for voice input instruction, on the display 121.

  After displaying the voice order input screen 60, the CPU 111 determines whether or not the voice recognition end is declared in ST13. When the voice input end is not declared, the CPU 111 determines whether or not the speech key 67 is turned on in ST14. If the speech key 67 is not turned on, the CPU 111 returns to the process of ST13.

  When detecting that the utterance key 67 is touched by a signal from the touch panel sensor 122 in ST14, the CPU 111 receives the voice data input through the microphone 13 and converted into digital data by the A / D converter 118. The voice recognition engine 119 is taken in. The capturing of the voice data continues until the finger is released from the speech key 67 and turned off in ST16. When detecting the turn-off operation of the utterance key 67, the CPU 111 operates the voice recognition engine 119 in ST17 to recognize the voice pattern of the voice data captured during the period when the utterance key 67 is turned on.

  Here, the CPU 111 constitutes an input section detecting means for detecting the start and end of voice input by the processes of ST14 and ST16. Further, the processing of ST15 constitutes a voice recognition means for recognizing a voice pattern input via the voice input means within a section from the voice input start to the end.

  When the voice pattern recognition of the input voice data is completed, the CPU 111 executes a menu candidate search process specifically shown in FIG. 6 as ST18. That is, the CPU 111 overwrites the input sound data memory 41 with the input sound data corresponding to the sound pattern recognized as ST31. In ST32, the menu item code, the item name, and the match rate data are cleared from the order candidate table 43, and the order candidate table 43 is initialized.

  Next, the CPU 111 reads the item name voice data of one menu product from the voice menu file 32 and overwrites the comparison voice data memory 42 in ST33. In ST34, the input voice data in the input voice data memory 41 is compared with the item name voice data in the comparison voice data memory 42, and the voice pattern matching rate is calculated.

  Next, in step ST35, the CPU 111 determines whether or not the voice pattern matching rate exceeds a preset threshold value P (for example, 50%). If the coincidence rate is equal to or less than the threshold value P, the CPU 111 proceeds to the process of ST38.

  On the other hand, when the matching rate exceeds the threshold value P, the CPU 111 reads the menu item code and the item name corresponding to the item name audio data from the audio menu file 32 as ST36. Then, each data of the menu item code, item name, and matching rate is written in the order candidate table 43. Thereafter, when a plurality of data is stored in the order candidate table 43 as ST37, the data is rearranged so that the matching rate decreases in order from the data of the table number “1”. Thereafter, the process proceeds to ST38.

  In ST38, it is determined whether or not data of another menu product exists in the voice menu file 32. If other menu product data exists, the process returns to ST33, and the item name voice data of the other menu product is read from the voice menu file 32 and overwritten in the comparison voice data memory. Thereafter, the processes of ST33 to ST38 are executed again.

  Here, the CPU 111 obtains the target name voice data (item name voice data) for each target stored in the storage means (voice menu file 32) as the recognized voice pattern data by the processes of ST31 to ST35 and ST38. Similar name search means for searching and extracting a plurality of similar target name speech data is configured. Further, similar data acquisition means for acquiring the target code (product code) and the target name (item name) respectively corresponding to the plurality of extracted target name voice data by the processing of ST36 from the storage means (voice menu file 32). Configure.

  If it is confirmed in ST38 that there is no other menu product data in the voice menu file 32, the CPU 111 determines whether the number of data stored in the order candidate table 43 is 2 or more in ST39. . That is, it is determined whether or not two or more item name voice data having a matching rate between the input voice data and the voice pattern exceeding P% are extracted. If the number of data is 1 or less, the CPU 111 multiplies the threshold value P preset in ST40 by a coefficient less than 1 (for example, 0.9) to lower the threshold value P. . Thereafter, the process returns to ST32, the order candidate table 43 is initialized, and the processes after ST33 are executed again.

  When the number of data stored in the order candidate table 43 in ST39 is 2 or more, the CPU 111 stores two or more item name data stored in the order candidate table 43 in ST41 on the voice order input screen 60. The candidate menu keys 62 are displayed in order. This is the end of the menu candidate search process.

  Here, the CPU 111 and the display unit 12 with a touch panel constitute output means for selectively outputting a plurality of item names obtained by the similar data acquisition means.

  When the menu candidate search process is completed, the CPU 111 waits for the cancel key 65 of the voice order input screen 60 to be touched in ST19 or for selection of one of the candidate menu products in ST20. Here, when it is detected from the signal from the touch panel sensor 122 that the cancel key 65 is touched, the CPU 111 returns to the process of ST12. That is, the item name displayed on each candidate menu key 62 on the voice order input screen 60 is deleted. And the process after ST13 is performed again.

  On the other hand, when it is detected by a signal from the touch panel sensor 122 that any candidate menu key 62 displaying the item name is touched, the CPU 111 selects a menu corresponding to the item name data as ST21. The product code is read from the order candidate table 43. The menu item code and the order number n are registered in the order memory of the RAM 113. In addition, the item name and the order score n of the menu product are displayed in the order menu column 68. The order number n is set to “1” as an initial value, and when a multiplier is input by the ten key 61 and the multiplication key 63, the order point is the multiplier. Thereafter, the CPU 111 returns to the process of ST12 and deletes the item name from each candidate menu key 62 of the voice order input screen 60. And the process after ST13 is performed again.

  Here, when an arbitrary target name is selected from a plurality of output target names (item names) in the process of ST21, the CPU 111 reads the target names together with the target names from the storage means (voice menu file 32). A voice input data processing means for processing the target code (product code) as input data is configured.

  If it is detected in ST13 that the voice recognition end key 66 has been touched, the CPU 111 returns to the process of ST3 and switches the screen of the display 121 to the order input screen 50.

  In the restaurant using the order input device 1 of the present embodiment configured as described above, it is possible not only to input the contents of the customer's order by the operation of the touch key as usual, but also to input by voice. .

  When using order data input by voice, the operator touches the voice recognition key 57 on the order input screen 50 shown in FIG. Then, the voice order input screen 60 shown in FIG. 8 is displayed, and the operator speaks the item name of the menu item to be ordered while touching the speech key 67. At this time, it is not necessary to utter the item name accurately and carefully one tone at a time, and only a part of the item name may be uttered. The operator may be a store clerk or a customer.

  For example, suppose that the user speaks “ice” while touching the speech key 67 to order the menu item “ice cream”. Then, the input voice data captured and recognized from the microphone 13 between the time when the finger touches the speech key 67 and the time when the finger is released, and the item name voice data of various menu products registered in the voice menu file 32 in advance. Are individually compared, and a coincidence rate is calculated for each. Then, a plurality of menu products whose matching rate exceeds a predetermined threshold value P are extracted, and their item names are displayed on the plurality of candidate menu keys 62 on the voice order input screen 60, respectively.

  Menu items with a matching rate exceeding the threshold P for the content “ice” uttered now are “rice”, “lychee”, “half-rice”, “plum wine”, “ice cream” in descending order of matching rate. ”And“ Apple pie ”. In this case, the item names of the six items are displayed on the candidate menu keys 62 of the voice order input screen 60, as shown in FIG.

  The operator then touches the candidate menu key 62 on which the item name of the menu item “Ice Cream” to be ordered is displayed. Then, the menu item code and the order score of this menu item “ice cream” are registered in the order memory. Also, as shown in FIG. 10, the item name, the order score, and the amount are displayed in the order menu column 68 of the voice order input screen 60.

  In this way, the operator can input the order contents using the voice. If the desired item name is not displayed on the candidate menu key 62 after the voice is input, the voice may be input again after touching the cancel key 65.

  When the voice order data input is thus completed, the operator touches the voice recognition end key 66. Then, the screen returns to the order input screen 50, and the confirmation key 55 is touched. Then, since the screen is switched to the order confirmation screen, the operator confirms the order contents, and if there is no mistake, touches the confirmation key. Thus, the current order data is transmitted via the network and processed at the station.

  As described above, according to the present embodiment, the utterance key 67 is arranged on the voice order input screen 60 displayed on the touch panel type display 121, and the microphone 13 is used only during the period during which the utterance key 67 is touched. The input voice data is captured and recognized. Therefore, even if another conversation is made around the order input device 1 or the surrounding noise is loud, the sound immediately before the finger touches the speech key 67 and the sound immediately after the finger is released are recognized. Since it can be excluded from the voice data, the recognition rate can be increased. In addition, even when uttering an item name that includes a sound or sound repellent such as “n” or “tsu”, it is not recognized that the sound has ended before the sound or sound repellent. Can also increase the recognition rate.

  In addition, in the present embodiment, the input voice data recognized at a high recognition rate as described above is obtained for the matching rate with the item name voice data of various menu products registered in advance in the voice menu file 32, All menu items exceeding the threshold value P are extracted as candidate menu items. In addition, if one menu item whose matching rate exceeds the threshold value P cannot be extracted, or if only one item can be extracted, the threshold value P is set to a lower value and the search is performed again. Candidate menu products are extracted. Each item name of the extracted plurality of candidate menu products is displayed on the plurality of candidate menu keys 62 of the voice order input screen 60 so as to be selected, and the item name selected by the touch operation of the candidate menu key 62 is selected. The menu item code for the menu item is processed as order input data.

  Therefore, even if a part of the item name is uttered, the item name of the menu product ordered with a high probability can be displayed on the candidate menu key 62, so that the item name of the menu product is accurately and carefully uttered one by one. There is no need to do so, and voice input can be easily utilized. Further, after the voice input, the order input of the menu product can be performed simply by touching the candidate menu key 62 displaying the item name of the menu product to be ordered from among the item names displayed on the candidate menu key 62. Therefore, even in restaurants with a large number of menu items, it is possible to easily find and input touch keys of desired order items, which is efficient.

  The present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage.

  For example, in the above-described embodiment, the present invention is applied to the data input device that processes the order data input of the menu product ordered by the customer at the restaurant. However, the present invention can also be applied to voice data input other than the order data. it can. For example, when inputting destination data by voice to a ticket vending machine that sells tickets such as railways, or by ordering data to an information terminal that sells content such as tickets for concerts, movies, etc. and game software. It can also be applied to input. That is, the target name and the target code are not limited to the item name and the product code in the above embodiment.

  In the above embodiment, the function program for carrying out the invention is recorded in advance in the ROM 112 inside the apparatus. However, the present invention is not limited to this, and a similar function program may be downloaded from the network to the apparatus. A similar function program stored in a recording medium may be installed in the apparatus. The recording medium may be any form as long as the recording medium can store the program and can be read by the apparatus, such as a CD-ROM. In addition, the function obtained by installing or downloading in advance may be realized in cooperation with an OS (operating system) inside the apparatus.

  In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, the constituent elements over different embodiments may be combined.

1 is an external perspective view of an order input device according to an embodiment of the present invention. The block diagram which shows the principal part structure of the order input device. The schematic diagram which shows the main data files preserve | saved at HDD of the same order input device. The schematic diagram which shows the main memory areas formed in RAM of the same order input device. The flowchart which shows the principal part of the program processing procedure which concerns on this invention which CPU of the same order input device performs. 6 is a flowchart specifically showing a procedure of menu candidate search processing in FIG. 5. The schematic diagram which shows an example of the order input screen displayed on the display of the order input device. The schematic diagram which shows an example of the audio | voice order input screen displayed on the display of the same order input device. The schematic diagram which shows the example of a screen after the speech recognition of the voice order input screen. The schematic diagram which shows the example of a screen after order input of the voice order input screen.

  DESCRIPTION OF SYMBOLS 1 ... Order input device, 11 ... Apparatus main body, 12 ... Display unit, 13 ... Microphone, 31 ... Menu master file, 32 ... Voice menu file, 33 ... Acoustic dictionary file, 41 ... Input voice memory, 42 ... Contrast voice memory, 43 ... Order candidate table, 111 ... CPU, 119 ... Voice recognition engine, 121 ... Display, 122 ... Touch panel sensor.

Claims (6)

An audio menu file for storing an item name for identifying each of a plurality of items to be ordered, an item code set corresponding to the item name, and audio data of the item name;
Voice input means for inputting voice;
An utterance key that is continuously operated while inputting voice by the voice input means;
An input section detecting means for detecting a voice input section while the speech key is continuously operated ;
A speech recognition means for recognizing the speech pattern input through the voice input means to said speech input in the interval detected by the input section detecting means,
The voice pattern data recognized by the voice recognition means is individually compared with the item name voice data for each order target item stored in the voice menu file to obtain a match rate, and the match rate is preset. Similar name search means for extracting a plurality of item name voice data exceeding the threshold ,
And similar data acquisition means for acquiring the article code and item names corresponding to a plurality of item name audio data extracted by said similar name retrieval means from said voice menu file,
A candidate menu display means for selectably list item name of each of the candidate in decreasing order of the matching rate for the plurality of items name candidates obtained by the similar data acquisition unit,
When an arbitrary item name is selected from a plurality of item names displayed by the candidate menu display means, an item code read from the voice menu file based on the selected item name is used as order input data. Voice input data processing means for processing;
Order menu display means for displaying, as an order menu, a list of order input data including item names and order points processed as order input data by the voice input data processing means;
A confirmation key for confirming the order input data displayed in a list on the order menu display means;
Transmitting means for transmitting order input data including an item code and an order score for the item name displayed on the order menu display means to the station that stores and manages the order data when the confirmation key is operated ;
A data input device comprising: It has a display with a touch panel,
2. The data according to claim 1, wherein the candidate menu display means is a means for displaying a plurality of touch key screens respectively assigned with a plurality of item names obtained by the similar data acquisition means on the display. Input device. Screen control means for displaying a touch key image for voice input instruction on the display;
The input section detecting means detects when the finger touches the touch key image for voice input instruction as the start of voice input, and detects when the finger is released as the end of voice input. Item 3. The data input device according to Item 2. The similar name search means compares the voice pattern data recognized by the voice recognition means with the item name voice data for each object stored in the voice menu file to obtain a match rate, and the match rate is 2. The data input according to claim 1, wherein when extracting item name voice data having a predetermined threshold value or more, when a plurality of item name voice data cannot be extracted, the threshold value is lowered and the search is performed again. apparatus. A voice menu file for storing the audio data of a plurality of item name that identifies ordering goods for each and the item name goods code set corresponding to the item name, an audio input means for inputting a voice, the A computer data input method comprising: an utterance key that is continuously operated while a voice is input by a voice input means; and a voice recognition means that recognizes a voice input via the voice input means,
The computer detects a voice input section while the utterance key is continuously operated , recognizes a voice pattern input in the detected voice input section , and uses the recognized voice pattern data as the voice pattern data. The item name voice data for each order target item stored in the voice menu file is individually compared to obtain a match rate, and a plurality of item name voice data whose match rate exceeds a preset threshold are extracted. The product code and the item name respectively corresponding to the extracted plurality of item name voice data are obtained from the voice menu file, and the item name of each candidate is obtained in descending order of the match rate for the obtained item name candidates. selectably list display, when any item name from among a plurality of line items that are the list is selected, the audio menu file based on the selected item name and the Processing the goods code read out al as order entry data, displays a list of the order input data including the item name and the order number as the order menu, when the confirmation key is operated, displayed in the order menu display means A data input method characterized by transmitting order input data including an item code and an order score for an item name to a station that stores and manages the order data . A voice menu file for storing the audio data of a plurality of item name that identifies ordering goods for each and the item name goods code set corresponding to the item name, an audio input means for inputting a voice, the While inputting voice by voice input means, a computer equipped with an utterance key that is operated continuously ,
An input section detection function for detecting as a voice input section while the speech key is continuously operated ;
A voice recognition function for recognizing a voice pattern input via the voice input means in the voice input section detected by the input section detection means ;
The matching rate is obtained by comparing the voice pattern data recognized by the voice recognition means with the item name voice data for each order target item stored in the voice menu file , and the matching rate is preset. Similar name search function to extract multiple item name voice data exceeding the threshold ,
A similar data acquisition function for acquiring item codes and item names respectively corresponding to a plurality of item name audio data extracted by the similar name search function, from the audio menu file ;
A candidate menu display function for displaying a list of the item names of the candidates for the plurality of item name candidates obtained by the similar data acquisition function so that they can be selected in descending order of the matching rate ;
If any item name is selected from among a plurality of item name displayed by the candidate menu display function, the goods code read from said voice menu file based on the selected item name as order entry data The voice input data processing function to process,
An order menu display function for displaying, as an order menu, a list of order input data including item names and order points processed by the voice input data processing function;
Wherein when the order menu enter key to confirm the order entry data listed in the display function is operated, the order input data including the article code and the order number for the item name displayed in the order menu display function, order data A transmission function for transmitting to a station that stores and manages
A program to realize

Images