KR102333819B1 - Address voice input system - Google Patents

Abstract

The present invention relates to an address voice input system, and a general voice recognition server such as Google or Microsoft by voice input of an address, such as a logistics address including courier delivery requiring address input, a destination of a navigation device, etc. into an address voice input terminal After receiving the voice recognition result using the voice recognition technology of By correcting the voice input system, it is not only possible to correct accurately regardless of the voice inputter's vocalization or input tendency, but also corrects errors caused by the address input method to ensure accuracy, thereby improving the accuracy, unity and convenience of address address input. it's about

Description

Address voice input system

The present invention relates to an address voice input system, and a general voice recognition server such as Google or Microsoft by voice input of an address, such as a logistics address including courier delivery requiring address input, a destination of a navigation device, etc. into an address voice input terminal After receiving the voice recognition result using the voice recognition technology of By correcting the voice input system, it is not only possible to correct accurately regardless of the voice inputter's vocalization or input tendency, but also corrects errors caused by the address input method to ensure accuracy, thereby improving the accuracy, unity and convenience of address address input. it's about

In Korea, the address system was the only address system in which the land number assigned to each division, that is, the lot number as the address, was the only address system. With the full implementation of the address, most of the addresses in public law relations were replaced with street name addresses. However, in the private sector, even as of 2019, the two address systems are being mixed, and the reality is that an appropriate response is required.

Here, in Article 2 of the 'Road Name and Address Act', the street name address is stipulated as "address indicated by road name, building number, and detailed address (applicable only if there is a detailed address)".

The use of such an address system is mainly used to input addresses in application fields such as logistics and vehicle navigation represented by postal mail and courier service. In , a method of directly selecting desired data from a search result based on a road name, a building name, etc. is being applied.

If the input time for each address data can be reduced in the main application fields of such an address system, the efficiency of the overall work can be increased. However, according to the current address and postal code input system, as detailed search conditions or more search terms are set, the number of database search results can be reduced, and as the search conditions or search words are set simply, the number of search results increases.

As described above, it is insufficient to prepare a method to reduce the time to set search conditions or search terms and to reduce both the time to select the desired address data in the search results. A new address input method was required to allow selection within.

As one of the prior art techniques for improving this, "a method and apparatus for simplifying address input" of Korean Patent Laid-Open No. 10-2018-0073012 has been proposed, which will be described with reference to FIG. 1 as follows.

1 is a flowchart for explaining an example of a method for entering a street name address using an intermediate search result that is a search result using a dictionary corresponding to a search word type of the prior art.

Here, the street name address input system in step S410 may wait for the user to set a search word. The user's search term setting can include both a method in which the user directly inputs a search term, a method in which a user selects a search term from candidates suggested by the system, and the like. For example, after 1 second to 3 seconds) has elapsed, or when a user issues an input command (eg, enter key input), the next step can be progressed.

In step S420, the street name address input system may determine which type of search word set by the user is the first type, the second type, or the third type. For example, the first type, the second type, and the third type may correspond to a street name, a postal code, and a building name, respectively.

Here, the search word type may be determined by the street name address input system itself recognizing the search word type based on a direct setting by the user or the format of the search word set by the user, or the search word input by the user is a street name, zip code, or building name. It may have a complex type in which a combined search term is set.

In this case, the user or street name address input system may separate the search words by type, and the separated search words (ie, partial search words) may branch into a process according to each type. may branch to step S510, in the case of the second type of search word, branch to step S610, and in the case of the third type of search word, branch to step S710.

In step S430, the street name address input system may determine whether there is a selection by the user from among the branched and extracted intermediate search results.

If it is determined that the user does not select any intermediate results, the system may proceed to step S440 to output a search failure guidance message to the user, and then proceed to step S410 to enter the standby state.

If the user selects one or more of the intermediate search results, the process proceeds to step S450 to search for the original street name address record desired by the user in the address database.

In step S430, the user may select any one of the intermediate search results based on the search term, or may select one intermediate search result from each of two or more types of intermediate search results.

In step S450, the street name address input system may attempt to search the address database for original street name address data corresponding to the intermediate search result selected by the user.

If the search is successful in step S460, the process may proceed to step S470 to return the final search result.

According to the operation of the street name address input system according to the prior art, even when the user sets only one type of search word, the user is allowed to select any one of the intermediate search results, and corresponding to the selected intermediate search result By outputting only the final search results (that is, candidates belonging to the group of intermediate search results among street name addresses corresponding to the original record in the address database) to the user, the user's choices can be greatly reduced, thereby increasing the efficiency of the user entering the street name address. there was.

However, such a road name address input system according to the prior art is not only being applied to narrow application fields such as the use of street name addresses, but also has recently been improved in voice recognition technology to receive guidance such as application fields such as vehicle navigation. There is an increasing need to input a destination using voice recognition and thereby set a destination. In addition, it is required to improve the address input method in the fields of postal, courier and logistics, which require more accuracy in address input. have.

In addition, when the address input system is applied in the same field as well as when the application fields are different, it is difficult to secure the versatility of the address database because each address database is constructed.

The present invention devised to solve the problems of the prior art as described above is by voice inputting an address such as a logistics address including courier delivery requiring address input, a destination of a navigation device, etc. into an address voice input terminal such as Google or Microsoft The purpose of this is to promote convenience of address input by receiving a voice recognition result using the voice recognition technology of a general voice recognition server of

In addition, the present invention relates to an address voice input system, and by voice inputting an address such as a logistics address including courier delivery requiring address input, a destination of a navigation device, etc. into an address voice input terminal, a general voice such as Google or Microsoft Another object of the present invention is to secure accuracy for inputting an address by receiving a voice recognition result using a recognition technology and correcting a voice recognition error that may occur in the voice recognition result.

In addition, there is another object of the address voice input terminal to ensure unity according to the use of the address database used for address input by correcting the address to be input compared with the standard address database.

In addition, there is another object for enabling the spread of street name addresses to be used in a short period of time by using the lot address, street name address, and building name as addresses used for address input.

The above object of the present invention is to provide an address voice input system, comprising: an app driving step of driving an app for address input in an address input terminal; a voice input step of voice inputting the address of the address to be input into the app of the address input terminal; a voice recognition step of recognizing a voice input to an app of the address input terminal and returning a voice recognition result to the address input terminal; If the voice recognition result does not match the format of the standard address database, the first correction is pre-corrected to match, the address and details returned by comparing the form of administrative dong name with numbers in the lot address or the real address with the standard address database A second correction that combines addresses and a correction that matches the input address input with one of the plurality of return results when multiple results are returned for one address input, one address input that does not contain an actual address In case there is no result returned by comparing with the standard address database for 3 A calibration step in which any one or more calibrations are performed; and an address input step of allowing the address corrected in the calibration step to be input to the app of the address input terminal.

Therefore, the address voice input system of the present invention provides the convenience of address input by inputting the address input by voice into the address voice input terminal and receiving the voice recognition result using the voice recognition technology of a general voice recognition server such as Google or Microsoft. There is an achievable effect.

In addition, by receiving a voice recognition result using a general voice recognition technology and correcting a voice recognition error that may occur in the voice recognition result and correcting an error caused by a mixture of address input methods, it is possible to secure accuracy for address address input. There is another effect.

In addition, there is another effect for ensuring unity according to the use of the address database used for address address input by correcting the address to be input by comparing it with the standard address database in the address voice input terminal.

In addition, there is another effect for allowing the spread of street name addresses to be used in a short period of time by using the lot address, the street name address, and the building name as addresses used for address input.

1 is a flowchart for explaining an example of a road name address input method using an intermediate search result according to the prior art;
2 is a schematic diagram showing an embodiment of an address voice input system according to the present invention.

The terms or words used in the present specification and claims should not be construed as being limited to their ordinary or dictionary meanings, and the inventor may properly define the concept of the term in order to best describe his invention. It should be interpreted as meaning and concept consistent with the technical idea of the present invention based on the principle that there is.

Therefore, the configuration shown in the embodiments and drawings described in the present specification is only the most preferred embodiment of the present invention and does not represent all of the technical spirit of the present invention, so at the time of the present application, various It should be understood that there may be equivalents and variations.

In the present invention, administrative districts such as 'Jung-gu, Seoul' or 'Gangnam-gu, Seoul', which are administrative districts commonly included in the lot address and street name address, are unified as administrative district names, and since the technical gist of the present invention and the relevance are low, this It is assumed that the description is mostly omitted.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

2 is a schematic diagram showing an embodiment of an address voice input system according to the present invention. As shown in FIG. 2 , an app for inputting an address in the address input terminal is operated in the address input terminal for an address such as a logistics address including courier delivery requiring address input and a destination of the navigation device (S10).

Here, it is preferable that the address input terminal is a mobile terminal capable of voice input using a smart phone, a navigation device, or the like.

Thereafter, an address of an address to be input, such as a delivery delivery address or a vehicle operation destination, is voice-inputted into the app of the address input terminal by a courier worker or a user of a navigation device, etc. (S20).

Here, it is preferable that the address inputted by the user's voice is input completely, but when the purpose can be achieved only by partial input, it is more preferable in consideration of the convenience of input in many cases.

These addresses have a system of "administrative district, actual address (in case of street name address, street name and building number), detailed address". Therefore, voice input is optional, and the address of the 'administrative district' is the address up to the city, county, and gu of the address.

Thereafter, the voice recognition result is returned to the address input terminal for address input by recognizing the user's voice input to the app of the address input terminal (S30).

The user's voice recognition can also recognize the user's voice directly from an address input terminal implemented in a smartphone or navigation device, but since the device possessed by the user has limitations in scalability, companies such as Google or Microsoft that have already been commercialized It is more preferable to recognize the voice input by the user using the voice recognition technology of the voice recognition server that is already commercialized in , and receive it as text data and utilize the result.

If the required address, such as a logistics address including courier delivery that requires address input, or an address such as a destination of a navigation device, is completely returned as a result of such voice recognition, no separate operation may be required, but this Address correction is necessary in order to ensure that the address is completely completed due to errors in voice recognition related to input and partial address input to improve convenience when inputting an address.

This address correction is a first correction (S41) of pre-correcting to match if the voice recognition result does not match the format of the standard address database, the administrative dong name format including numbers in the lot address, or the real address with the standard address database A second correction (S42) that performs a correction combining the returned address and the detailed address by comparison, and correction of matching one of the plurality of returned results with the input address input when a plurality of results are returned for one address input; If there is no result returned by comparing with the standard address database for one address input that does not include a real address Address correction of any one or more of the third correction (S43) for performing the correction related to the building name ex post is performed (S40).

This address correction is made based on a standard address database compared with the voice recognition result, and it is more preferable that the standard address database is an address database of the Ministry of Public Administration and Security.

The first correction corresponds to correction of correcting the voice recognition result in the standard address database format in advance when the voice recognition result is input and does not match the format comparable to the standard address database that returns the address (S41).

As an example of this, if there is a space such as "△△△1 road" in the street name, the space is removed with "△△△1 road" There is a correction to remove the space as “ΔΔ Building” or to delete the number in the case of “ΔΔ 1-dong” that contains a number in the name of the administrative dong of the lot address.

In addition, in the case of a street address among the lot numbers, if "123-4" is voiced in, the result of voice recognition is displayed as "120 3-4". By correcting "123", errors can be reduced by correcting the address only with the main number of the street address based on the '-', which is the delimiter that distinguishes the main and minor numbers of the address among the lot numbers.

The second correction corresponds to a correction combining the returned address and the detailed address by comparing the real address with the standard address database (S42).

For example, if the real address, lot number address, or street name address and detailed address are combined and entered, it corresponds to a correction in which the returned address is compared with the standard address database and the detailed address, such as "Δ△ Building 3F Team", etc. is combined. do.

In this case, the detailed address is combined after being stored in the temporary memory of the address input terminal. Here, when the building name is included in the detailed address, it is also possible to classify the building name by the brand name and the building properties, and a separate building name database is constructed in the address voice input system.

Here, the building property means an apartment, a building, a tower, a school, a villa, or a center.

The third correction is a correction that matches one of the plurality of return results with the input address input when multiple results are returned for one address input, and compares one address input without an actual address with a standard address database If there is no result returned by adding the administrative dong name of the actual address, the correction is returned, and if the returned result includes the building name, it corresponds to the correction related to the building name ex post facto (S43).

In this example, there is no change in the administrative dong for the input of one address "△△ dong 3", but for '3', there is a plurality of results such as '1-3', '2-3', '3', 4-3' is returned from the standard address database and calibrated to an exact matching address.

Here, since it is difficult to compare matching addresses when the number of the plurality of results is relatively large, the number is limited to a certain number that is easy to compare, and the addresses are directly compared to find and correct the exact addresses.

In addition, in the case of correction combining the building name and the detailed address, if the returned address includes the building name compared to the standard address database, it may overlap with the building name included in the detailed address. do.

Also, in most cases, users are not required to enter all of the address system when they are in charge of landmark functions such as large apartment complexes and community centers, and therefore, when entering the address, "△ △ apartment + number of houses" Alternatively, it is often used to enter only the brand name of a building, such as “△△3 building + 3rd floor of community center”, which is the building attribute, and the detailed address. Perform a correction that combines addresses.

Here, the brand name or building attribute of the building constituting the building name is preferably constructed in the address voice input system as a building name database, and it is more preferable that spaces are excluded from the building name.

Taking this as an example of "△△3 building + 3rd floor of community center", in case of "△△3 building + 3rd floor of community center", the voice recognition result of the voice recognition server is "△△3 building" + "community center" Each is searched separately like "+"3rd floor", and when comparing this with the standard address database, the real address cannot be found in most cases.

Here, it is preferable that the brand name of the building or the property of the building is constructed in the address voice input system as a building name database.

Here, since the address including the building name is returned by input of the lot address or street name address, it is duplicated with the building name of the detailed address, so it is preferable to correct the duplication of the building name by deleting it so that it is composed of only one building name.

The address corrected in the address correction step (S40) is inputted to the app of the address input terminal (S50).

Although the present invention has been illustrated and described with reference to preferred embodiments as described above, it is not limited to the above-described embodiments, and those of ordinary skill in the art to which the present invention pertains within the scope not departing from the spirit of the present invention Various changes and modifications will be possible.

Claims (15)

In the address voice input system,
App driving step of driving an app for address input in the address input terminal (S10);
a voice input step (S20) in which the address of the address to be input by the user is voice input to the app of the address input terminal;
A voice recognition step (S30) of recognizing a voice input to the app of the address input terminal and returning a voice recognition result to the address input terminal;
If the voice recognition result does not match the format of the standard address database, the first correction (S41) for pre-correcting to match, the form of administrative dong name containing numbers in the lot address, or the real address returned by comparing the standard address database The second correction (S42) for combining the address and the detailed address, and if a plurality of results are returned for one address input, one of the plurality of returned results and correction to match the input address input, including the actual address If there is no result returned by comparing with the standard address database for a single address input that has not been entered, a correction to receive the result by adding the administrative dong name of the real address is returned. address correction step (S40) in which any one or more corrections are performed among the third corrections (S43) for performing corrections related to ; and
It consists of an address input step (S50) of allowing the address corrected in the address correction step to be input to the app of the address input terminal,
The second correction is a correction for combining the returned address and the detailed address by comparing the real address with the standard address database when the real address and the detailed address are combined and input.
According to claim 1,
Address voice input system, characterized in that the address to be inputted by voice optionally includes an administrative district address.
3. The method of claim 2,
The address of the administrative district optionally included is an address voice input system, characterized in that the address up to the city, county, district.
According to claim 1,
The voice recognition is an address voice input system, characterized in that it is performed in a commercialized external voice recognition server.
According to claim 1,
The voice recognition result is text data.
According to claim 1,
The address voice input system, characterized in that the standard address database is an address database of the Ministry of Public Administration and Security.
According to claim 1,
In the first correction, it is determined whether the voice recognition result returned from the voice recognition step matches the format of a standard address database to be compared, and if it does not match the format of the standard address database, the first correction is corrected in the format of the standard address database. .
8. The method of claim 7,
The first correction is to remove a space from a street name or a building name among detailed addresses when the voice recognition result does not conform to the format of the standard address database, or to delete a number when the administrative dong name of the lot number address contains a number Address voice input system, characterized in that the correction.
7. The method of claim 6,
The first correction is an address voice input system, characterized in that it is a correction for correcting the address only with the original number of the street number based on '-', which is a separator that separates the main number and the minor number of the street number among addresses.
delete According to claim 1,
Address voice input system, characterized in that when the detailed address includes the building name, the building name is divided into a brand name and a building attribute.
According to claim 1,
When the third correction is a correction combining the building name and the detailed address, it is compared with the standard address database, and any one of the building name included in the returned address and the building name included in the detailed address is deleted. Address voice input system.
According to claim 1,
The third correction is an address voice input system, characterized in that when input is only the brand name of the building, the building name and the detailed address that are the building attributes, the correction combining the administrative dong and the brand name of the building, the building attributes and the detailed address in the real address .
14. The method of claim 11 or 13,
The address voice input system, characterized in that the brand name or building attribute of the building is built into the address voice input system as a building name database.
According to claim 1,
In the third correction, when a plurality of results are returned for one address input, the number returned is set to a certain number and received, and the returned value and the input value are directly compared and matched. If the registered building name is not returned from the address database of the Ministry of Public Administration and Security, the building name of the detailed address in the case where the address containing the building name is returned by input of the correction to return the address by additionally combining the real address Address voice input system, characterized in that it is a correction that removes duplication so that it is made with only one building name because it is duplicated.

Images