JPH09258653A - Map recognition processing system and map recognition processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable the easy execution of setting operations at the time of building a system and to substantially prevent the occurrence of setting errors at the time of building the system by constituting the system so as to have an information setting means, speech recognizing means, facility extracting means and correspondence processing execution means. SOLUTION: The system consisting of maps as a base is provided with various kinds of map background informations 2, facility information 4 and a recognition processing table 8 associated in correspondence thereto as a framework for building the system, by which the need for labor at the time of building the system is eliminated and the operation time is shortened for the higher efficiency of the operation. In addition, the manual operations are decreased and, therefore, the building in of the human errors is lessened. Building of the various map recognition processing system is made possible simply by changing the setting of the map background information 2, a facility information table 4, a facility data base 6, the recognition processing table 8, a recognition dictionary table 9 and a processing program correspondence table 17.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a map recognition processing system and a map recognition processing method, and more particularly to a map recognition processing system and a search system utilizing a voice recognition response device for providing information such as a tourist guide. It can also be applied to an information input system that can input information to various facilities and arbitrary points. Especially, even when the number of facilities of the map to be voice-recognized is large, the setting at the time of system construction is possible. The present invention relates to a map recognition processing system and a map recognition processing method capable of easily performing work.

[0002]

2. Description of the Related Art FIG. 16 is a flow chart showing a processing flow of a conventional map recognition processing system, and FIG. 17 is a view showing a recognition dictionary table used in the conventional map recognition processing system. The recognition dictionary table shown in FIG.
The reading of the place name written on the map and the corresponding recognition ID are shown. When the reading of the place name on the map is “Shinjuku”, the recognition ID is “1”, and when the reading of the place name on the map is “Shibuya”, the recognition ID is “2”, and the recognition ID is on the map. When the reading of the place name is "Shinagawa", the recognition I
D is "3".

Next, the operation of the conventional map recognition processing system will be described. First, the recognition dictionary table is set in advance so that the recognition dictionary table shown in FIG. 17 is used as the recognition dictionary (step S1001). Next, waiting for voice input waiting for recognition processing (step S100
When the voice "Shibuya" is input in the state of 2), the recognition dictionary table is searched, the corresponding recognition word is searched, and the recognition ID "2" that matches the reading "Shibuya" is RI.
Return to D (return ID) variable (step S100)
3). Here, if recognized correctly, the recognition ID "2" of the voice "Shibuya" is obtained as a recognition result (step S1004). Finally, the street introduction video of Shibuya, which is the processing corresponding to the recognition ID “2”, is reproduced (step S1
005).

Similarly, when the voice "SHINJUKU" is input in the state of waiting for voice input waiting for recognition processing (step S1002) after setting the recognition dictionary table, the recognition dictionary table is searched and the corresponding recognition word is searched. (Step S1
003). Then, when the recognition ID "1" of "shinjuku" is obtained as a recognition result (step S100).
4), the street introduction video of Shinjuku, which is the processing corresponding to the recognition ID “1”, is reproduced (step S1006).

Similarly, after setting the recognition dictionary table, waiting for voice input waiting for recognition processing (step S100).
When the voice "Shinagawa" is input in the state of 2), the recognition dictionary table is searched to find the corresponding recognition word (step S1003). Then, the recognition ID "3" of "Shinagawa" is obtained as a recognition result (step S100).
4), the street introduction video of Shinagawa, which is the processing corresponding to the recognition ID “3”, is reproduced (step S1007).

When a voice other than "Shinjuku", "Shibuya", and "Shinagawa" is input in the state of waiting for voice input waiting for recognition processing (step S1002), the recognition dictionary table is searched and the corresponding The recognized word is checked (step S1003). If the corresponding recognition ID is not obtained as the recognition result (step S1004), the process returns to the voice input waiting (step S1002).

In this conventional map recognition processing system, when any one of "Shinjuku", "Shibuya", and "Shinagawa" is input by voice, each voice is recognized and the street corresponding to each voice is recognized. It has an advantage that the introduction video can be automatically reproduced.

[0008]

However, in the above-mentioned conventional map recognition processing system, the facility name of the map to be voice-recognized is the main function,
The setting process for recognizing the facility name of the map must be built into the system by C program one by one, and the system must be constructed. When the number of facilities for voice recognition increases, the setting work becomes enormous. Therefore, there is a problem that the setting work at the time of constructing the system becomes difficult and a setting error is likely to occur at the time of constructing the system.

Further, in the above-mentioned conventional map recognition processing system, since the setting process for recognizing the facility name on the map is constructed as a system, if the setting process for recognizing the facility name on the map is desired to be changed, the setting process is changed. There was a problem that it was difficult to do.

Therefore, according to the present invention, even if the number of facilities of the map to be voice-recognized is large, the setting work at the time of system construction can be easily performed, and the setting error can be made less likely to occur at the time of system construction. It is an object of the present invention to provide a map recognition processing system which can be changed and can be easily changed when the setting processing at the time of system construction is desired to be changed.

[0011]

A first invention is a facility information table for setting map background information, facility information for displaying various facilities displayed on a map, and a facility corresponding to the facility information table. Recognition processing table for setting facility readings for causing the voice recognition processing means to recognize, a recognition dictionary table for setting recognition words for causing the voice recognition processing means to recognize, and a recognition ID to be returned when recognized, and at the time of voice recognition It is provided with a processing program correspondence table that sets the correspondence processing to be executed, map background information, facility information table,
An information setting means for setting various kinds of information of the recognition processing table, the recognition dictionary table and the processing program correspondence table, a map background information / facility information display means for displaying map background information and facility information, and recognizes the inputted voice, A voice recognition processing unit that extracts a recognition ID based on the voice recognition result, a facility extraction unit that extracts a facility number corresponding to the recognition ID from the recognition processing table, and a facility number extracted based on the processing program correspondence table. The present invention is characterized by having a corresponding process executing means for executing a corresponding process for a facility.

A second aspect of the invention is a map conversion processing means for generating map background information, a facility information table and a facility database based on existing map information, and a recognition dictionary conversion for generating a recognition processing table based on the facility information table. It has a processing means.

A third aspect of the invention is characterized by having a reading conversion processing means for generating the facility reading of the recognition processing table based on the facility name of the facility information table.

A fourth aspect of the invention is an arbitrary point data designation processing means for designating an arbitrary point on a map, an information input processing means for inputting information, and an empty facility extraction processing for extracting a vacant facility number from a facility information table. It is characterized by comprising means and arbitrary data registration processing means for allocating the extraction facility number to a designated arbitrary point and registering the designated arbitrary point information in the facility information table, the facility database, and the recognition processing table.

A fifth aspect of the invention is characterized by having facility number reading automatic addition processing means for automatically adding the reading of the facility number when performing the recognition processing edit processing for generating the recognition dictionary table from the recognition processing table. To do.

[0016] A sixth aspect of the present invention is a facility information table for setting map background information and facility information for displaying various facilities displayed on a map, and a facility for voice recognition processing means corresponding to the facility information table. A recognition processing table for setting facility readings for recognition, a recognition dictionary table for setting recognition words for recognition by the voice recognition processing means and a recognition ID to be returned when recognized, and corresponding processing executed at the time of voice recognition are set. And a process program correspondence table for setting various information of the map background information, facility information table, recognition processing table, recognition dictionary table and process program correspondence table, and then displaying the map background information and facility information. Then, the step of recognizing the input voice, and then extracting the facility number based on the voice recognition result. And a step of extracting a facility number corresponding to the recognition ID from the recognition processing table, and a step of performing a corresponding process for the facility of the facility number extracted based on the processing program correspondence table. It is what

A seventh aspect of the invention is characterized in that after the map background information, the facility information table and the facility database are generated based on the existing map information, the recognition processing table is generated based on the facility information table. is there.

The eighth invention is characterized in that the facility reading of the recognition processing table is generated based on the facility name of the facility information table.

The ninth aspect of the invention is to specify an arbitrary point on the map, input information, extract an empty facility number from the facility information table, and set the extracted facility number to the specified arbitrary point. Allocation, facility information table, facility database, and a step of registering designated arbitrary point information in the recognition processing table.

A tenth aspect of the invention is characterized by having a step of automatically adding reading of the facility number when the recognition dictionary table is generated from the recognition processing table.

[0021]

Embodiments of the present invention will be described below with reference to the drawings. Embodiment 1. FIG. 1 is a block diagram showing a configuration of a map recognition processing system according to an embodiment of the present invention, FIG. 2 is a diagram showing an example of map background information generated by a map edit processing unit shown in FIG. 1, and FIG. 1 is a diagram showing an example of a facility information table set by the facility information edit processing unit shown in FIG.
5 is a diagram showing an example of a facility database set by the facility database editing processing unit shown in FIG. 5, FIG. 5 is a diagram showing an example of a recognition processing table set by the recognition processing editing processing unit shown in FIG. 1, and FIG. 6 is a recognition shown in FIG. FIG. 7 is a diagram showing an example of a recognition dictionary table created by the recognition processing edit processing unit based on the processing table, and FIG. 7 is a diagram showing an example of a processing program correspondence table set by the processing program correspondence table editing processing unit shown in FIG. .

In FIG. 1, 1 is a map edit processing unit for generating map background information 2, and 3 is a facility information edit processing unit for setting a facility information table 4. The facility information table 4 sets facility information for displaying various facilities displayed on the map. Reference numeral 5 is a facility database edit processing unit that sets the facility database 6, and reference numeral 7 is a recognition processing edit processing unit that sets the recognition processing table 8 and creates a recognition dictionary table 9 based on the recognition processing table 8. Reference numeral 16 is a processing program correspondence table editing unit for setting the processing program correspondence table 17.

Corresponding to the facility information table 4, the recognition processing table 8 sets facility readings for causing the facility extraction processing unit 19 to extract the facility, and the recognition dictionary table 9 causes the voice recognition processing unit 10 to recognize. A recognition word for and a recognition ID to be returned when recognized are set. Voice recognition processing unit 1
0 recognizes the voice input by the voice input unit 11 such as a microphone and a handset, and recognizes I based on the voice recognition result.
Extract D. The facility extraction processing unit 19 extracts the facility number corresponding to the recognition ID from the recognition processing table 8.

Reference numeral 12 is a map display processing unit for displaying the map background information 2 and facility information on the display 13.
Is a correspondence processing unit that executes a correspondence process corresponding to the facility of the facility number extracted by the voice recognition processing unit 10 based on the processing program correspondence table shown in FIG. 7, and 15 is the voice recognition processing unit 10 and the map display processing. It is a system integrated processing unit that manages the unit 12, the correspondence processing unit 14, and the like.

Next, the operation of the map recognition processing system of this embodiment will be described. In this embodiment, a case of constructing a map recognition processing system will be described. First, the map edit processing unit 1 generates the map background information 2 shown in FIG. 2 by raster information. In the present embodiment, the map background information 2 is created by the map editing processing unit 1 with raster information represented by a bitmap, but it may be created with vector information.

As shown in FIG. 3, the facility information edit processing unit 3 generates a facility information table 4 for setting facility information for displaying various facilities displayed on the map. The facility information table 4 shown in FIG. 3 corresponds to each facility number, a facility name including information such as the A library and the B library corresponding to each facility number, a facility position for displaying each facility, and each facility. And a display pattern. The facility position is indicated by the bit of the raster information, and indicates the display position of the x and y axes of the bit information.

The facility database edit processing section 5 sets the facility database 6 as shown in FIG. The facility database 6 shown in FIG. 4 includes a facility number, a facility name corresponding to each facility number, and a library of attribute information corresponding to each facility,
It is composed of a facility type including information such as an art museum, and an introduction video file name corresponding to each facility. The facility number and facility name in the facility information table 4 correspond to the facility number and facility name in the facility database 6, respectively.

In the present embodiment, it is possible to set the facility type of the attribute information and the introduction video file by the facility information guidance. The name of the introduction video file can be displayed in a list, and a relation can be set with the facility. Further, by selecting a facility on the map screen, the introduction video file associated with the above can be reproduced.

As shown in FIG. 5, the recognition processing edit processing unit 7 sets the facility reading for causing the voice recognition processing unit 10 to recognize the facility. In the conventional system, the display process and the recognition process are performed separately, and therefore, they have to be created separately. In addition, it was necessary to identify the correspondence with the ID and associate it. On the other hand, in the present embodiment, the display process, the recognition process, and the association thereof are performed by the tool of the present system, so that the reading on the setting screen can be input without mistake.

As shown in FIG. 6, the recognition processing edit processing section 7 sets a recognition dictionary for setting a recognition word to be recognized by the voice recognition processing section 10 and a recognition ID returned when the recognition is performed, as shown in FIG. Set Table 9. Based on the processing program table 17, the processing program correspondence table edit processing unit 16 sets an execution program for a designated facility for each corresponding processing number. By switching the setting of the corresponding process number before executing the program, a plurality of processes can be switched and executed.

Next, FIG. 8 is a diagram showing a processing flow when the map recognition processing system constructed with the above various settings is executed. The system control processing unit 15 sets a corresponding processing number that determines a processing program executed at a designated time as an initial setting (step S1). Here, it is assumed that 1 is set. The map display processing unit 12 reads the map background information 2 set by the map editing processing unit 1 shown in FIG. 2 and displays the map background on the display display 13 (step S2). Next, the facility information table 4 is read, and the facility information is displayed on the map background based on the facility position information and the display information of FIG. The voice recognition processing unit 10 sets the recognition dictionary table 9 shown in FIG. 6 related to the displayed map as a recognition processing target (step S4). The pre-processing of the voice recognition is completed up to this point, and the voice input is awaited (step S
5).

The voice recognition processing unit 10 receives the voice "Dae-no-dojo" from the voice input unit 11,
The inputted voice information is converted into character information and collated with the recognition dictionary table shown in FIG. 6, and the matched recognition ID “4” is R.
Return to ID (step S6). The facility extraction processing unit 19
The facility number 4 of the record having the same RID and recognition ID is acquired from the recognition processing table 8 shown in FIG. 5 (step S
7). The correspondence processing unit 14 obtains the video reproduction program from the processing program correspondence table 17 as the processing program corresponding to the corresponding processing number 1 set in step S1 (step S9). The video reproduction program executes reproduction of the introduction video file “introduction 0004.mpg” of the facility number 4 from the facility database 6 of FIG. 4 (step S10). Similarly, the introduction video information can be reproduced for the A library, B library, and C library by speaking the corresponding readings as in the D playground. Further, by changing the handling process, it is possible to build an information box-like system such as inputting opinions to the facility and inputting contact matters.

Conventionally, it was necessary to describe the explanation of the map and the corresponding processing by a program, but in the present embodiment, various map background information as a framework for constructing a map-based system, By providing the facility information and the recognition processing table 8 associated with them, it is possible to save time and effort when constructing the system, and it is possible to shorten the work time and improve the work efficiency. . In addition, since human labor can be reduced, human error can be reduced. Further, the system can be flexibly dealt with by changing the map, changing the recognition dictionary, etc. instead of changing at the program level.
Further, various map recognition processing systems can be constructed only by changing the settings of the map background information 2, the facility information table 4, the facility database 6, the recognition processing table 8, the recognition dictionary table 9, and the processing program correspondence table 17. .

Embodiment 2 Next, an embodiment of automatically generating map background information, a facility information table, and a recognition processing table in the case of constructing a recognition processing system regarding a map will be described below with reference to the drawings. FIG. 9 is a diagram showing an automatic generation processing configuration for automatically generating map background information, facility information table and recognition processing table according to the second embodiment of the present invention. 9, the same reference numerals as those in FIG. 1 indicate the same or corresponding portions, and 21 indicates the map background information 2, the facility information table 4, based on the existing map information 22.
A map conversion processing unit that generates the facility database 6,
23 is a recognition dictionary conversion processing unit that generates the recognition processing table 8 based on the facility information table 4, and 24 is a reading conversion processing unit that generates the facility reading of the recognition processing table 8 based on the facility name of the facility information table 4. Is.

In the present embodiment, since various digital map formats such as the map information of the Geographical Survey Institute currently exist, the existing map information 22 is utilized. This allows
Since the map background information 2, the facility information table 4, and the facility database 6 can be automatically generated, mistakes can be further reduced as compared with the case of the manual input of the first embodiment.

The map conversion processing unit 21 uses the existing map information 2
2 is used to convert the existing map information 22 into raster information. Here, the map background information 2 is automatically set by designating the range, the resolution, the magnification, and the like. The map conversion processing unit 21 automatically sets the facility information table 4 by using the existing map information 22 having the attribute information and selecting and extracting the attribute information such as a library. Map conversion processing unit 21
Sets the facility database 6 similarly using the existing map information 21. In the present embodiment, the map background information 2 is generated by the map conversion processing unit 21 as raster information represented by a bitmap, but it may be generated by vector information.

The reading conversion processing unit 24 automatically sets the facility reading of the recognition processing table 8 based on the facility name of the facility information table 4. If the existing map format has no reading item, the reading conversion processing unit 24 can be used to set the automatically converted reading information. When the facility information table 4 set by the facility information edit processing unit 3 in FIG. 1 exists, the recognition dictionary conversion processing unit 23 can automatically convert the contents of the facility information table 4 into the recognition processing table 8. it can. At this time, there is no reading information,
The reading conversion processing unit 24 can automatically set reading items.

In the present embodiment, the background information of the map and the facility information can be simplified by designating the range and the like by using the digitized data of the existing map and the type of the facility. Can be automatically generated. Also,
It is possible to efficiently automate the generation of data related to the field of such a system.

In the present embodiment, by adding a process of converting display information including kanji of reading into reading information of only hiragana, not only information about the map but also reading information about the recognition process is automatically generated. Therefore, the system can be constructed more easily.

Embodiment 3 FIG. 10 is a block diagram showing the configuration of the map recognition processing system according to the third embodiment of the present invention. 10, the same reference numerals as those in FIG. 1 indicate the same or corresponding portions, and 30 is an arbitrary point data information input section for inputting information at an arbitrary position.

Since the facility information table 4, the facility database 6, the recognition processing table 8 and the recognition dictionary table 9 are set in advance, when the registration processing is performed, the registration position is limited to the set facility. On the other hand, the present embodiment is effective in the case of inputting at an arbitrary point, for example, in the case of inputting information such as that the side of the pedestrian bridge is broken and the repair is required quickly.

Next, an example of arbitrary point data input processing of the present embodiment will be described with reference to FIG. First, the input position on the map displayed on the map display processing unit 12 is designated (step S101). Here, (200,12
It is assumed that the position of 0) is designated. Next, the information desired to be registered at the designated position is input (step S102). The information to be input may be characters / figures / still images / moving images, but here, moving image data is input and tmp00
5. It is assumed that a file called avi has been created. The facility information table is searched so that the arbitrary point input above can be treated like other facility information, a free facility number is input (step S103), and the facility number is assigned to the designated arbitrary point.

Here, if the facility information table is in the state of FIG. 3, 5 is assigned as the facility number of the designated arbitrary point. After that, the facility information table (step S
104), the facility database (step S105), and the recognition dictionary table (step S106) to register the input information. When each table of FIGS. 3 to 5 is registered before processing, and as shown in FIGS. 12 to 14, each table is updated. Finally, the recognition dictionary table is regenerated based on the contents of the recognition dictionary table, and the recognition is restarted (step S107).

In the present embodiment, a free facility number is acquired from the facility information table, and an arbitrary point data input processing step for inputting information to the facility having the acquired facility number is added to set at the time of system construction. Not only fixed facility information but also data can be registered and displayed as information on any point, so it is possible to input detailed data from the viewpoint of the actual system user. it can. Therefore, the flexibility of the system can be increased and the usability can be improved.

In addition, when the recognition processing edit processing unit 7 generates the recognition dictionary table 9 from the recognition processing table 8, the facility number reading automatic addition processing unit 7a is executed, and the reading of the facility number is fixed in the recognition dictionary. By adding this, the program can be continuously executed without changing the recognition dictionary table 9. As shown in the recognition dictionary table 9 in FIG. 15, for example, the recognition dictionary table 9 is created by adding and fixing information so that up to 100 facilities can be read. And
When the facility number is displayed at the time of the display process and, for example, "Good evening" is pronounced, the video corresponding to it is reproduced. Also, for example, when pronounced "Ichiban",
"Etoshokan" is selected and the corresponding video is played.

In the present embodiment, since it is not necessary to change the recognition dictionary table 9 when inputting arbitrary point data, the processing can be continuously executed without interrupting the speech recognition processing, and the speech is continuously recognized. When recognizing, it is possible to minimize the deterioration of response.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a map recognition processing system according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an example of map background information generated by a map edit processing unit shown in FIG.

3 is a diagram showing an example of a facility information table set by a facility information edit processing unit shown in FIG.

FIG. 4 is a diagram showing an example of a facility database set by a facility database edit processing unit shown in FIG.

5 is a diagram showing an example of a recognition processing table set by a recognition processing edit processing unit shown in FIG.

FIG. 6 is a diagram showing an example of a recognition dictionary table created by a recognition processing edit processing unit based on the recognition processing table shown in FIG.

FIG. 7 is a diagram showing an example of a correspondence processing program table when executing a correspondence processing for a facility having an extracted facility number.

FIG. 8 is a flowchart showing a processing flow when the map recognition processing system shown in FIG. 1 is executed.

FIG. 9 is map background information according to the second embodiment of the present invention,
It is a figure which shows the automatic generation process structure which performs automatic generation processing of a facility information table and a facility database.

FIG. 10 is a block diagram showing a configuration of a map recognition processing system according to a third embodiment of the present invention.

11 is a flowchart showing a processing flow of an arbitrary data input processing unit in FIG.

12 is a diagram showing an example of a facility information table added to FIG. 3 by the arbitrary data input processing unit shown in FIG.

13 is a diagram showing an example of a facility database added to FIG. 4 by the arbitrary data input processing unit shown in FIG.

14 is a diagram showing an example of a recognition dictionary table added to FIG. 5 by the arbitrary data input processing unit shown in FIG.

FIG. 15 is a diagram showing an example of a recognition dictionary table generated by the facility number reading automatic addition processing unit shown in FIG.

FIG. 16 is a flowchart showing a processing flow of a conventional map recognition processing system.

FIG. 17 is a diagram showing a recognition dictionary table used in a conventional map recognition processing system.

[Explanation of symbols]

1 map editing processing unit, 2 map background information, 3 facility information editing processing unit, 4 facility information table, 5 facility database editing processing unit, 6 facility database, 7 recognition processing editing processing unit, 7a facility number reading automatic addition processing unit, 8 recognition processing table, 9 recognition dictionary table, 10 voice recognition processing section, 11 voice input section, 12 map display processing section, 1
3 display, 14 corresponding processing unit, 14a arbitrary point data input processing unit, 15 system control processing unit,
16 processing program correspondence table edit processing unit, 17
Processing program correspondence table, 18 processing program section, 19 facility extraction processing section, 21 map conversion processing section, 2
2 existing map information, 23 recognition dictionary conversion processing unit, 24
Reading conversion processing unit, 30 arbitrary point data input processing unit.

Claims (10)

[Claims] 1. A facility information table for setting map background information, facility information for displaying various facilities displayed on the map, and a facility for causing the voice recognition processing means to recognize the facility in correspondence with the facility information table. A recognition processing table for setting facility readings, a recognition dictionary table for setting a recognition word to be recognized by the voice recognition processing means and a recognition ID to be returned when recognized, and a processing program for setting corresponding processing executed at the time of voice recognition And a table background information, facility information table, recognition processing table, recognition dictionary table, and information setting means for setting various information of the processing program correspondence table, and map background information / facility displaying the map background information and facility information The information display means and the input voice are recognized, and the recognition I is performed based on the voice recognition result.
A voice recognition processing means for extracting D, a facility extraction means for extracting a facility number corresponding to the recognition ID from the recognition processing table, and a correspondence processing for executing the correspondence processing for the facility of the facility number extracted based on the processing program correspondence table. A map recognition processing system comprising: a processing execution means. 2. A map conversion processing means for generating map background information, a facility information table and a facility database based on existing map information, and a recognition dictionary conversion processing means for generating a recognition processing table based on the facility information table. The map recognition processing system according to claim 1, further comprising: 3. The map recognition processing system according to claim 2, further comprising reading conversion processing means for generating a facility reading of the recognition processing table based on the facility name of the facility information table. 4. An arbitrary point data designation processing means for designating an arbitrary point on a map, an information input processing means for inputting information, and an empty facility extraction processing means for extracting an empty facility number from a facility information table. The map recognition process according to claim 1, further comprising: a facility number assigned to a designated arbitrary point, and a facility information table, a facility database, and an arbitrary data registration processing unit that registers the designated arbitrary point information in a recognition processing table. system. 5. The facility number reading automatic addition processing means for automatically adding the reading of the facility number when performing the recognition processing edit processing for generating the recognition dictionary table from the recognition processing table. The map recognition processing system described in. 6. A facility information table for setting map background information, facility information for displaying various facilities displayed on a map, and a facility for causing voice recognition processing means to recognize the facility in correspondence with the facility information table. A recognition processing table for setting facility readings, a recognition dictionary table for setting a recognition word to be recognized by the voice recognition processing means and a recognition ID to be returned when recognized, and a processing program for setting corresponding processing executed at the time of voice recognition A step of setting various information of a map background information, a facility information table, a recognition processing table, a recognition dictionary table and a processing program correspondence table, and a step of displaying the map background information and the facility information, and then Recognizing the input voice, and then extracting a facility number based on the voice recognition result, Next, the map recognition is characterized by including a step of extracting a facility number corresponding to the recognition ID from the recognition processing table and a step of performing a corresponding process for the facility of the facility number extracted based on the processing program correspondence table. Processing method. 7. The recognition processing table according to claim 6, wherein after the map background information, the facility information table and the facility database are generated based on the existing map information, the recognition processing table is generated based on the facility information table. Map recognition processing method. 8. The map recognition processing method according to claim 7, wherein the facility reading of the recognition processing table is generated based on the facility name of the facility information table. 9. A step of designating an arbitrary point on the map, a step of inputting information, a step of extracting a vacant facility number from a facility information table, and assigning the extracted facility number to the designated arbitrary point, and facility information. The map recognition processing method according to claim 6, further comprising a step of registering designated arbitrary point information in a table, a facility database, and a recognition processing table. 10. The map recognition processing method according to claim 9, further comprising the step of automatically adding the reading of the facility number when the recognition dictionary table is generated from the recognition processing table.