JP2003004470A - Voice recognition unit for navigation - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently perform the voice recognition processing operation of a place name or an installation name, to reduce the processing burden of a voice recognition unit and to enhance processing speed. SOLUTION: The voice recognition dictionaries of Japan, the U.S.A. or the like as a whole are divided into latitudes and longitudes so as to be stored in a plurality of latitude-and-longitude division-block unit voice recognition dictionary storages 13. A present-position signal is input from a navigation system 28, a block to which a present position belongs is selected, a peripheral block which exists in the circumference of the block is selected, a latitude and longitude division block which corresponds to a group of the blocks is selected, and a present-position-computing-block group voice recognition dictionary 14 is created. When a user inputs a voice from a microphone 1, a voice recognition part 4 retrieves the phoneme data in the dictionary 14, and the place name or the facility name which corresponds to the input voice is retrieved. When a plurality of candidates exist, they are narrowed down by a plurality-of-candidates narrow-down processing part 20, by using categories or distances from the present position.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition device for a navigation device, which enables a user to recognize a voice input from a microphone and set a destination of the navigation device and various searches. The present invention relates to a navigation voice recognition device provided with a voice recognition dictionary that enables high-speed retrieval of data corresponding to voices such as place names and facilities input by a person.

[0002]

2. Description of the Related Art In a conventional navigation device 30 as shown in FIG. 15, for example, map data for drawing a map and map / information data 3 recording various information are recorded.
3. CD-RO which records various data such as place name facility name corresponding point data 34 which records latitude / longitude data of a point corresponding to the inputted place name or facility name as described later.
A map / information storage medium 32 such as an M or a DVD-ROM is provided. Of the data in this map / information storage medium 32,
Map data centered on a point designated by the system control unit 31 is read out by the map drawing unit 35, and is combined with other various images by the image combining device 36, and the image display device 3 is displayed.
In FIG. 7, a map and information related thereto are drawn in various modes such as 3D.

The system control unit 31 includes a manual operation input unit 4
0 is connected and inputs an operation signal from a manual operation unit operated by a user such as a remote controller, a key switch, and a touch panel. Further, due to recent advances in voice recognition technology, a voice recognition device has been incorporated in a navigation device. In the navigation device 30 of FIG. 15, a voice recognition device 50 that recognizes a user's voice from a microphone 51. Is equipped with. As a result, by performing a voice recognition process as described later, the voice of the user recognized from the recognition result output unit 58 is output to the system control unit 31 of the navigation device 30, and the user operates the same as the manual operation input unit 40. It can be used as an instruction signal.

In response to the various operation instruction signals from the user, the destination and waypoint setting unit 44 sets the point designated by the user as the destination or waypoint. As for the current location, for example, a GPS signal is used, and an accurate current position is detected using a vehicle speed sensor or a traveling direction sensor,
This map data centering on the present location is read from the map data recorded in the map / information storage medium 32,
By displaying this on the image display device 37 and displaying the current position of the vehicle in an overlapping manner, it is possible to see at a glance where the vehicle is currently traveling.

The system control unit 31 includes a guide route calculation unit 4
2 is connected and the link data recorded in the map / information storage medium 32 is used based on the present location, the destination and the transit point as described above, and the time and distance are included in the route connecting these points. , The most suitable route is automatically searched in consideration of conditions such as charges, the searched route is set as a guide route, and this guide route is displayed on the screen by changing the color of the road on the map image and drawing it thick I am able to do so.

Further, in order to allow the driver to travel reliably along the guide route set as described above, in the guide route guide section 43, for example, an intersection where the vehicle should change the route on the guide route. When approaching within a certain distance, the intersection is magnified and displayed on the screen by drawing an arrow, etc. showing the direction to change the course, or by voice guidance to turn the user to the destination. I am able to guide you to.

[0007]

Various means have been conventionally used for designating the above-mentioned destination or transit point, or for searching a specific point and displaying it on the screen. Data that is listed in order from the wide area side to the narrow area side is stored in advance in the map / information storage medium 32, and the user can use the on-screen keyboard or the like in order from the prefecture in the same manner as a normal address display. A method of finally narrowing down a specific place name is also adopted by specifying the place name by inputting it or by selecting place names sequentially displayed on the screen from the wide area side by inputting a cursor instruction or the like.

[0008] In addition, the map / information storage medium 32 includes public facilities such as stations and city halls, road facilities such as intersections and interchanges, and landmarks such as buildings, convenience stores, gas stations, and restaurants. It is also possible to narrow down a specific facility by storing it and allowing the user to directly input this facility name or display a candidate list and select it by inputting a cursor instruction or the like. The specific place name or specific facility name set in this way is input to the search place name facility name input unit 45 in FIG. 15, and the point data search unit 46 indicates the latitude / longitude corresponding to the place name or facility name. The point data is searched from the place name facility name corresponding point data 34 of the map information storage medium 32.
This data is used as point data of a destination or a waypoint, or is data representing this point on the map screen.

It is not preferable for the driver to manually search for the place name or facility name as described above while the vehicle is running. As a countermeasure, the voice recognition device 5 as described above
As described above, the voice recognition dictionary unit 55 is provided with a place name facility name dictionary unit 56, and when a place name or facility name is input from the microphone 51 by the user, the place name or facility name is input from the voice input unit 52. Input to the parameter extraction unit 53,
The characteristic parameter of the voice obtained here is output to the voice recognition unit 54, and the voice recognition unit 54 searches the place name facility name dictionary unit 56 for one having a parameter most suitable for this parameter, and retrieves this. The recognition result output unit 58 outputs the search site name facility name input unit 45. The place name and the facility name by the voice recognition input here are output to the spot data searching unit 46 in the same manner as the manual input by the user, and the spot data corresponding to the specific place name and the facility name is obtained in the same manner as above. Used for setting destinations and displaying on the map.

As described above, the place name or facility name can be directly searched by voice, but when the voice is input, the place names can be selected by uttering the place names in order from the prefecture in the same manner as the manual input. It is also possible to narrow down the voice recognition processing load. However, in that case, it is troublesome because the voice input must be performed many times. Therefore, it is possible to set this only by uttering the most detailed place name part, and when there are multiple points, the candidate points are displayed in a list or displayed on a map. Is preferably performed.

To that end, when searching for a specific place name or facility name such as "Yoshima" as the place name and "Twin Tower" as the facility name, the place name facility name dictionary unit 5 is used.
It is necessary to search from all the place names and facility names in Japan recorded in 6. Retrieving one place name or facility name from a huge amount of data in this way imposes a heavy load on the processing device that performs this arithmetic processing, and the processing time becomes longer and the recognition rate is inevitably reduced. If a high-performance arithmetic processing device is used, it will be expensive.

As a countermeasure, the data of the place name facility name dictionary unit 56 is divided into predetermined regions such as prefecture units, and when searching for a specific place name or facility name, an input for specifying the region is made beforehand, or navigation is performed. It is also conceivable that the area including the point where the vehicle currently exists is specified by the data of the current position of the device, and only the voice recognition dictionary included in the specific area such as a prefecture unit is used for the voice recognition.

However, by only referencing the place name facility name dictionary for a specific area in this manner, for example, when a person at the prefectural border between Ibaraki and Fukushima travels between prefectures, this navigation device is used. When traveling over a region, such as when using, the voice recognition of a place name existing in an adjacent region is not performed, which makes the device extremely difficult to use.

As a countermeasure, it may be possible to solve the above-mentioned problem by selecting a voice recognition dictionary so as to include a prefecture adjacent to a prefecture including a point where a vehicle currently exists. At that time, for example, when the current location is Fukushima prefecture, Ibaraki prefecture, Tochigi prefecture, Niigata prefecture, Yamagata prefecture,
Miyagi Prefecture will be selected and an extremely large area will be selected. Also, for example, when the current location is in Tokyo, Kanagawa, Yamanashi, Saitama, and Chiba are selected.Since the area is relatively small but the population density is extremely high, the place name and facility name included in that area It becomes a large amount in proportion to it. Also, for example, when the current location is Shimane Prefecture, only Yamaguchi Prefecture, Hiroshima Prefecture, and Tottori Prefecture are selected, and a relatively narrow range is selected and the population density is also relatively low. The number of place names and facility names are relatively small.

In this way, the range covered by the voice recognition dictionary of place name facility names and the number of place name facility names greatly vary depending on the point where the vehicle currently exists. Therefore, in the voice recognition device, it may be possible to match it with an area having an intermediate amount of data, but in that case, when using this voice recognition device in a vehicle currently in Tokyo as described above, the processing is performed. The speed is inappropriately slowed down, which annoys the user. In addition, even when there are vehicles in Tokyo, when using a high-performance arithmetic processing unit with a processing capacity capable of performing accurate voice recognition processing at high speed, it is excessive in most other regions. Processing capacity of.

The above-mentioned problem is not limited to Japan, but is the same in the United States, for example, in order to recognize a place name in the United States by speech, for example, a speech recognition dictionary targeting an adjacent "State". In the case of setting, the number of place name facility names including the neighboring states and the area of the target region also differ greatly depending on the states, as in the case of the above-mentioned Japanese example.

Therefore, according to the present invention, when the voice recognition processing is performed using the voice recognition dictionary in which the place name and the facility name are set for each area, the voice recognition dictionary in an appropriate range is set and is appropriately selected. In view of the above, a main object is to provide a navigation voice recognition device that enables high-speed and accurate voice recognition using an inexpensive data processing device.

[0018]

In order to solve the above-mentioned problems, a navigation voice recognition apparatus according to the present invention provides phoneme data of place names and facility names included in a block divided by latitude and longitude and its related data, A block unit voice recognition dictionary storage unit having a plurality of block unit voice recognition dictionaries recorded for each block, a current position belonging block to which the current position belongs, and a peripheral block in a predetermined range around it are selected. A current position-corresponding block group speech recognition dictionary selected from the block unit speech recognition dictionary storage unit, the current position-corresponding block group selecting unit, and the block unit speech recognition dictionary of the block group selected by the current position-corresponding block group selecting unit And the phoneme data corresponding to the voice of the input place name or facility name is searched from the current position corresponding block group voice recognition dictionary. Those having a speech recognition processing unit that outputs.

Further, another navigation voice recognition apparatus according to the present invention is such that the blocks divided by the latitude and longitude are so large that the place names and the facility names included in each block are as equal as possible. It is divided into different sizes.

In another navigation voice recognition apparatus according to the present invention, the blocks around the block to which the current position belongs are blocks adjacent to the current position belonging block.

Further, another navigation voice recognition apparatus according to the present invention comprises: a block surrounding the block to which the current position belongs; a block adjacent to the block to which the current position belongs; and a block in a predetermined range around the block. It is included.

In another navigation voice recognition apparatus according to the present invention, the current position of the block around the block to which the current position belongs is moved so that the amounts of the place name and the facility name are as equal as possible. It was made to change with.

Further, in another voice recognition device for navigation according to the present invention, the voice recognition processing unit exists in a block close to the current position with respect to the place name or facility name included in the current position corresponding block group voice recognition dictionary. The higher the degree of speech approximation is set.

Further, in another voice recognition device for navigation according to the present invention, the current position corresponding block group voice recognition dictionary comprises a list in which blocks of the current position corresponding block group are recorded, and the voice recognition processing section, A place name or facility name having phoneme data corresponding to the input voice is searched from the dictionary data of the block recorded in the list of the block unit voice recognition dictionary storage unit, based on the list recording the block. It was done.

Further, in another navigation voice recognition device according to the present invention, the current position belonging block selection unit changes the block to which the current position belongs due to the movement of the current position and selects a new current position belonging block. At this time, peripheral block selection processing is performed.

Further, another navigation voice recognition device according to the present invention comprises a plurality of candidate narrowing down processing units, and the plurality of candidate narrowing down processing units perform voice recognition processing by the current position corresponding block group voice recognition dictionary. As a result, the narrowing-down process is performed when there are a plurality of candidates.

Further, another navigation voice recognition apparatus according to the present invention comprises a category-based voice recognition dictionary in which place names and facility names are recorded by category, and the plural candidate narrowing down processing unit is a category designated by a user. The voice recognition processing is performed by using the category-based voice recognition dictionary corresponding to.

Further, in another speech recognition apparatus for navigation according to the present invention, the plural candidate narrowing-down processing section calculates the distance from the current position for each of the plural candidates obtained as a result of the speech recognition processing, and the calculation result According to the output of the candidate point distance order arrangement unit, a plurality of candidates are sequentially displayed in a list on the display unit so that the user can make a selection.

Further, another navigation voice recognition apparatus according to the present invention calculates the distance from the current position for each of the plurality of candidates obtained as a result of the voice recognition processing, and outputs the candidate having the shortest distance as the voice recognition result. It was done like this.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram showing main functional units of a navigation voice recognition apparatus according to the present invention and their mutual relations. In the conventional voice recognition apparatus 50 shown in FIG. 15, the voice recognition processing of the present invention is performed. It shows the functional blocks to be performed. In the figure, the voice from the microphone 1 is input from the voice input unit 2, the feature of the voice input by the voice feature parameter extraction unit 3 is extracted, and is output to the voice recognition unit 4. The voice recognition unit 4 appropriately selects various dictionaries of the voice recognition dictionary unit 5 and searches the phoneme data of the selected dictionary for the one that matches the extracted feature parameter of the voice.

The voice recognition dictionary unit 5 is provided with a place name / facility name dictionary unit 6 similarly to the conventional place name / facility name dictionary unit 56 shown in FIG. 15, and a place name is set for setting a destination or a waypoint. This dictionary can be used to perform voice recognition processing when inputting, or when inputting a facility that serves as a guide when setting a destination or transit point, and when using a specific facility while the vehicle is traveling. I have to. In addition to such place names and facility names, various functions of the navigation device, such as enlargement or reduction of the scale of the displayed map and change of the map display such as 3D display, are operated by voice. The other dictionary 7 is provided. The selection of this dictionary is performed by determining whether or not the input voice includes a voice for operating a predetermined function of the navigation device by using another dictionary 7 such as an operation function, When it is not included, it can be used to determine that it is a voice related to a place name or a facility name.

As shown in FIG. 5A, the place name / facility name dictionary unit 6 has, as an example, a latitude division area divided by latitudes a1, a2, ... And longitudes b1, b2 ,. For the block of the area where the longitude division area is divided by, the voice recognition data of the place name and facility name included in this area and the position data of the point where the place name and facility name exist are recorded for voice recognition. It is provided with a latitude / longitude division block voice recognition dictionary 13 which is a dictionary. In the example shown in FIG. 5A, the block (n, m) to which the point where the current position is located belongs to the n area which is a latitude division area between latitudes a3 and a4, and between the longitudes b3 and b4. It is shown as a block of an area where the m area, which is a longitude division area, intersects.

In the example of FIG. 5A, n + 1, n + 2 and n-1, n- are centered on the n region which is the latitude division region.
2 and each of the areas m + 1, m + 2 and m-1, m-2 centering on the area m which is the longitude division area.
+ 2, m-2), (n + 1, m-2), (n, m-)
2) ... (n + 2, m-1), (n + 1, m-)
1), ... (N-2, m + 2) exist.

The place name facility name dictionary unit 6 of FIG. 1 outputs the current position signal of a functional unit for detecting the current position of the vehicle, such as the current position detecting unit 41 of FIG. It is input from the position input unit 10. The current position belonging block selection unit 11 stores the voice recognition dictionary stored in the latitude / longitude divided block voice recognition dictionary 13 according to the current position signal input from the current position input unit 10 and divided into each block as described above. Select the block to which the current position belongs from. Figure 5
In the example shown in (a), the current location of the vehicle exists in the block (n, m) as described above, and this block is selected.

When the current position affiliation block is selected in this way, the data is output to the peripheral block selection unit 12, and in the peripheral block selection unit 12, for example, all the surroundings adjacent to the block in which the current position exists are surrounded. A predetermined peripheral block is selected so that the above block is selected. FIG. 5B shows an example at that time,
Surrounding blocks (n + 1, m-1) and (n, m-) adjacent to the current position belonging block (n, m) as peripheral blocks
1), (n-1, m-1), (n + 1, m), (n-
1, m), (n + 1, m + 1), (n, m + 1), (n
A total of 8 blocks (-1, m + 1) are selected.

In FIG. 1, a block unit corresponding to a total of nine blocks, one block selected by the current position belonging block selection unit 11 and eight blocks selected by the peripheral block selection unit 12 as described above. The voice recognition dictionary is extracted and stored in the current position corresponding block group voice recognition dictionary 14. The voice recognition unit 4 performs voice recognition processing of a place name and a facility name using the current position belonging block group voice recognition dictionary 14.

The current position affiliation block selection unit 11 always detects whether or not the current position is in the range of the previously selected current position affiliation block according to the current position signal that changes with the movement of the vehicle. When moving from that block to another block, a new block to which the current position belongs is selected immediately. This state is shown in FIG. 5C, where the vehicle is at a point P on the road L of the block (n, m).
When passing through the boundary point Q between the block (n, m) and the block (n, m + 1), entering the block (n, m + 1), and traveling in the direction of the point R, at the time when the boundary point Q is passed, The position belonging block selection unit 11 detects that the current position belonging block is changed to the block (n, m + 1), and selects this block.

The selection result is output to the peripheral block selection unit 12, and the peripheral block selection unit 12 selects eight blocks around the current position belonging block in the same manner as described above. As a result of such selection, as shown in FIG. 5D, the current position belonging block (n, m + 1) and its surrounding (n + 1,
m), (n, m), (n-1, m), (n + 1, m +)
1), (n-1, m + 1), (n + 1, m + 2),
A total of 8 blocks of (n, m + 2) and (n + 1, m + 2) are selected.

The voice recognition dictionaries corresponding to the nine blocks selected as described above are accumulated in the current position block group voice recognition dictionary 14, and the user wants to know Seven-Eleven Koyama Otome store at the point Q, for example. At this time, when the voice "Seven-Eleven Koyama Otome store" is input from the microphone, the voice recognition unit 4 searches the facility name in the current position corresponding block group voice recognition dictionary 14 for the facility name close to the input voice, for example, "Seven-Eleven 〇". 〇 In addition to “stores”, it is possible to search for facility names that exist near the vehicle, such as “Seven Eagle” (sports store) and “Seven East” (travel store).

When setting a block divided by latitude and longitude as shown in FIG. 5A, the latitude and longitude can be set arbitrarily, but if this is set in a narrow area, The number of place names and facility names included in the block group will be small. Therefore, if a desired place name or facility name exists in this data, the total amount of data searched by the voice recognition unit 4 is small, and therefore voice recognition processing is performed at high speed even if a CPU having the same processing speed is used. be able to. However, since the range of the voice recognition target is limited, there is a high possibility that the block does not exist in the block group, and in that respect, the voice recognition processing is not properly performed.

On the other hand, when the setting of the blocks divided by latitude and longitude is set to a wide area, the number of place names and facility names included in the block group at that time increases, and the voice recognition unit 4 Since the total amount of data to be searched becomes large, it takes a lot of time to perform the voice recognition process when the CPUs having the same processing speed are used. However, at that time, the range of the voice recognition becomes wider, so there is a high possibility that the place name or point intended by the user exists in the block group, and in that respect, the voice recognition processing is properly performed. Becomes

As described above, when appropriately setting the range to be subjected to the voice recognition, for example, when setting the guide route, the voice input of the place name of the destination far from the present location and the guide route are performed. If you are looking for a nearby 7-Eleven to do shopping while traveling along,
Since the target area to be searched greatly differs, it may be set in advance so that the selection range of the peripheral block can be arbitrarily selected depending on the situation in which the voice recognition is performed. In that case, a selection range adjusting unit is provided in the peripheral block selecting unit 12 in FIG. 1 so that a user can arbitrarily select or input a voice for setting a guide route and a voice input while traveling along the guide route. May be set and the range may be automatically changed.

FIG. 6 (a) shows an example of adjusting the selection range of the peripheral blocks as described above. The present position is a latitude division area between latitudes a5 and a6, and longitudes are b5 and b.
When the longitude division area between 6 exists in the intersecting block S1, and the range of the peripheral block is set narrow, the longitude is included between latitudes a4 and a7 between b4 and b7.
When a block group S2 is selected as a peripheral block and the range of the peripheral block is set to be wide, latitudes a2 and a
For example, 49 block groups S3 having a longitude between b2 and b9 are set. In addition to always setting the same size in all areas, set such a block group to a narrow range in areas with many place names and facility names, and to set wide in areas with few area names. It can also be changed to.

Further, when setting a predetermined range of the voice recognition dictionary as described above, for example, as shown in FIG. 6A, a mesh of latitude and longitude is finely set, and the range is broadly divided into a total of 49 blocks. For the group S3 selected, a large latitude and longitude mesh is set as shown in FIG. 7B, and the current position belonging block T1 between the latitudes c3 and c4 with the longitude between d3 and d4 is set. The nine peripheral blocks can be selected as the substantially same speech recognition dictionary as that selected as the peripheral block group T2, and the speech recognition dictionary group can be arbitrarily selected by such adjustment.

As shown in FIG. 6 (a), when the block group S2 adjacent to the current position belonging block S1 and the block group S3 in the appropriately set range around the block group S2 are set, In addition to arbitrarily switching the block group selection as described above, the closer to the current position belonging block S1, the more the weighting coefficient of the proximity calculation processing with the input voice in the voice recognition processing is performed for the place name and facility name included in the block It can be set to be larger and to determine that the name intended by the user is highly likely. By making such settings, more accurate voice recognition is possible. For example, for all place names and facility names, the distance from the current position is calculated and weighting of each candidate is performed as described above. It is possible to further reduce the load of the voice recognition process by collectively performing the calculation for each block.

Further, when setting the range of the voice recognition dictionary, it is preferable to set the number of places of the place name or facility name included in each block within a predetermined range in consideration of the processing capacity of the voice recognition device. . At this time, in the block group obtained as a result of the block formation in this way, the block group having the largest number of points is set to a range which can be sufficiently dealt with by the voice recognition device.

When dealing with a block having a large number of place names or facility names, even a block having the largest place name or facility name can be sufficiently dealt with by the processing capacity of the speech recognition apparatus used. In the case of subdividing the size into the entire range, the number of places of the place name or the facility name is extremely small in other areas, which may not be preferable as the range of the voice recognition dictionary.

As a countermeasure, for example, as shown in FIG. 7A, a range A of a block including an area with many place names / facility names is subdivided and a latitude a3 for block division is set between latitudes a2 and a4. However, the latitude a5 for block division is set between the latitudes a4 and a6. Similarly, between the longitudes b3 and b5, b4 is set between the longitudes b3 and b5, and b6 is set between the longitudes b5 and b7, so that the range A of the block is 1/4 of the other blocks. Form blocks of area and equalize the number of places of place names and facility names included in one block as much as possible. When there is a further need, for example, as shown in FIG. 7B, in the range A of the block of FIG. 7A, particularly in the range B of the block having many place names / facility names, each block is divided into four sizes. It is also possible to form blocks and to equalize the number of place names / facility names included in each block.

By performing such fine division, for example, when forming a block for a voice recognition dictionary in which one country is divided by latitude and longitude, for example, in the Kanto region of Japan, the capital Tokyo. Since there are many place names and facility names in the area due to the high population density of the area where the capital is located, a subdivided block range A is formed in this area. Since there is a place name / facility name of, the area B is used by forming a range B of a block obtained by subdividing this part.

For example, when the block for the voice recognition dictionary is divided as shown in FIG. 7A, the current position correspondence as shown in FIGS. 8A, 8B, and 8C is accompanied by the movement of the vehicle. A block group is selected. That is, the current position exists in the block F1 portion in FIG. 8A, and the current position corresponding block group G1 including the block groups existing around the block is selected. At this time, the current position block F
Since 1 is adjacent to the block subdivision range, a narrow range is selected as a peripheral block in the block subdivision range portion.

Next, when the current position moves from the block F1 to the block F2 in the same figure (b), the current position corresponding block group G2 is selected. At this time, since the current position belonging block F2 is a block in the block subdivision range, the blocks around it are narrower. Further, when the current position moves to the block F3 as shown in FIG. 7C, all the surrounding blocks are block subdivision ranges, so that the current position corresponding block group G3 becomes the narrowest range. In this way, when subdividing a block with a large number of place names / facility names, it is possible to perform voice recognition processing while keeping the number of place names / facility names included in the voice recognition dictionary almost always constant. . The same operation is performed when the block is subdivided as shown in FIG.

The voice recognition unit 4 in FIG. 1 performs voice recognition of a place name or a facility name using the voice recognition dictionaries of various modes as described above. As a result of the recognition, the voice recognition unit 4 corresponds to the input voice. If there is only one place name or facility name to be output, this is directly output to the output confirmation unit 26, and the operation signal input unit 25
When the confirmation signal of the user from is input, it is output from the recognition result output unit 27 to the navigation device 28.

At this time, when there are a plurality of candidates for a place name or a facility name, in this embodiment, the data is output to a plurality of candidate list display output section 23 and the list is displayed on the display section 24 of the navigation device. In addition, the plurality of candidate lists are output to the category-based voice recognition unit 21 of the plurality of candidate narrowing-down processing unit 20, and when searching for a place name or a facility name intended by the user from among the candidates, the candidate-based voice recognition is performed. To be able to recognize. For this reason, in this embodiment, the place name / facility name dictionary unit 6 is provided with a place name / facility name category-based voice recognition dictionary 15 so that voice recognition can be performed based on the category. In the above example, when there are a plurality of candidates as a result of the recognition processing in the voice recognition unit 4, the plurality of candidates are displayed in the display unit 24 as a list, but without displaying them. , Prompts you to select a genre immediately,
The user may be prompted to select the genre.

The user refers to the list of categories separately displayed on the display unit 24 to select a desired category by operating the cursor keys of the remote controller in the operation signal input unit 25, or voice-inputs the category name separately. The voice recognition unit 21 outputs the signal to the category-based voice recognition unit 21. In addition, in the voice recognition process at this time, the dictionary 7 such as the operation function in the voice recognition dictionary unit 5 is used. Speech recognition unit by category 21
Searches the dictionary of the corresponding category of the place name / facility name category dictionary 15 according to the signal of the category designated by the user, and selects the phoneme data of the place name or facility name recorded in the category dictionary. , Search for and match the ones that match the voice input by the user.

By the above-described voice recognition processing in the category-based voice recognition unit 21, when the above-mentioned plurality of candidates are narrowed down, and as a result, there is only one candidate, this is directly passed to the output confirmation unit 26. It is output and the confirmation of the user is requested in the same manner as described above, and when confirmed, this is output from the recognition result output unit 27 to the navigation device 28. At this time, when a plurality of candidates still exist, the candidates are output to the candidate point distance order arrangement narrowing unit 22, the distance from the current position is calculated for each candidate, and a list arranged in order from the closest distance is created. The list is output to the display unit 24 via the plural candidate list display output unit 23 to display the list.

The user refers to this list display and selects an appropriate one from the operation signal input section 25 and outputs it to the candidate point distance order arrangement narrowing section 22. In the candidate point distance order arrangement narrowing unit 22, the designated candidates are output to the output confirmation unit 26, and the output confirmation unit 26 outputs the same to the display unit 24 in the same manner as described above, and confirmation from the user who saw this. When a signal is input from the operation signal input unit, the recognition result output unit 27
To the navigation device 28 via. In the above embodiment, an example in which a list in which the plurality of candidates are arranged in the order of distance is displayed is shown, but the closest candidate is output to the output confirmation unit 26 without requesting the list display to request confirmation from the user. You can also set

The navigation voice recognition apparatus shown in FIG. 1 and composed of the functional blocks for performing the above-mentioned functions can be operated in sequence by the operation flows shown in FIGS. 2 and 3, for example. Hereinafter, description will be given with reference to the functional block diagram of FIG. In the voice recognition process, a recognition voice is first input from the user as shown in FIG. 2 (step S1). Then, the feature parameter of the voice is extracted (step S2),
It is determined whether or not the input voice includes a voice regarding a place name or a facility name (step S3).

At the time of this determination, the voice input unit 2 of FIG.
The voice feature parameter extraction unit 3 extracts the feature parameter of the voice from the microphone 1 input in
Can be performed by searching and determining whether or not the dictionary 7 corresponding to the extracted voice feature parameter exists in the dictionary 7 such as the operation function in the voice recognition dictionary unit 5. This determination is made in the voice recognition dictionary unit 5,
The data of the place name / facility name dictionary unit 6 is, for example, millions of POIs.
There is a case where huge amount of data such as (Point of interest) exists, whereas it is usual that only a few hundred words exist in the dictionary 7 such as the operation function. By using the dictionary 7 for other functions and the like, the above discrimination can be easily performed. However, in addition to such a determination method, the input voice is determined to be a voice related to a place name / facility name by detecting that the voice input is during the user's destination setting input operation. When you recognize the word "near", which is a predetermined specific term when searching for a facility name such as "nearby convenience store", the next word is the facility name. Can also be determined as

In this determination, when it is determined that the voice uttered by the user includes the place name or the facility name, the process of setting the voice recognition dictionary of the place name / facility name corresponding to the current position is performed (step S4). The processing flow is shown in FIG. 3 and will be described in detail later, but in this step, as described above, in this step, the country or region targeted by this navigation device is divided into a plurality of blocks of regions divided by latitude and longitude. The included place name and facility name are formed as one unit voice recognition dictionary, and the place name / facility name voice recognition dictionary is formed as a set. Therefore, select the block in which the current location exists, and further The current position corresponding block group voice recognition dictionary is set by selecting blocks in a predetermined range around the block.

Using the voice recognition dictionary set in this way, a voice data search process corresponding to the input voice feature parameters is performed (step S5), and the place name and facility name corresponding to the voice data are obtained. Various necessary data such as position data of the place where they exist, which are recorded together with the place name and facility name, are read out. Since various retrieval methods conventionally used can be used for the retrieval processing of the voice recognition dictionary, description thereof will be omitted here.

As a result of the search processing performed as described above, it is determined whether or not there is only one point corresponding to the obtained place name or facility name (step S6). If it is determined that there is not only one, that is, there are a plurality of candidates, a process of narrowing down a plurality of candidates is performed (step S7). This processing flow is shown in FIG. 4 and will be described in detail later. As described above, the plural candidate narrowing processing unit 20 of FIG. 1 narrows down the categories, and further, if necessary, separates the plurality of candidate points. It arranges in order, and the narrowing down process is performed in which the user refers to this and selects.

After performing such a narrowing down process, the obtained voice recognition result is output to the display unit or the like to prompt the user for confirmation (step S8). Next, the user determines whether or not the recognition result is appropriate (step S9), and when it is determined that the recognition result is appropriate, it is determined whether or not the voice input is ended (step S10). When it is determined that the voice input is finished because the user does not input a new voice for a predetermined period thereafter, the operation flow is finished (step S13). If it is determined in step S6 that there is only one candidate as a result of the voice recognition dictionary search processing, the process immediately proceeds to step S8 to confirm whether the obtained candidate is intended by the user. To do.

On the other hand, when it is determined in step S3 that the voice input by the user does not include a voice relating to a place name or a facility name, a dictionary for performing voice recognition such as an operation function is selected. (Step S11),
The dictionary is searched for a word corresponding to the input voice. As for the words obtained by the voice recognition, similar to the above, this is displayed on the display unit, or output such as uttering a voice for inquiring whether to perform a specific function, etc., and confirming the user. Encourage.

When it is determined in step S10 that the voice input is not completed, as when the user again inputs voice, the process returns to step S1 and the same operation as described above is repeated. Further, in the step S9, the user makes a judgment on the confirmation output of the result of the voice recognition, and when it is judged that the recognition result is not the intention of the user, a signal to that effect is input to the step 1 Return to, input voice for recognition again,
The same operation is performed thereafter.

The setting process of the voice recognition dictionary of the place name / facility name corresponding to the current position in step S4 can be performed by the operation flow shown in FIG. 3, for example. At the beginning of this processing, data of the current position is fetched (step S21). This operation is performed by the current position input unit 10 in FIG. 1 capturing a signal from the current position detection unit of the navigation device 28.

In this embodiment, it is then determined whether the voice recognition dictionary selection process to be performed is the first one (step S22). If it is determined that the process is the first voice recognition dictionary selection processing, the current position belonging block is selected (step S23). This process is performed by the latitude / longitude division block unit voice recognition dictionary 1 in FIG.
3 includes, for example, a block portion formed at a portion where a latitude division area divided by latitude and a longitude division area divided by longitude as shown in FIG.
It can be obtained by forming a list to which an identification number is added in advance and selecting a specific block from the latitude and longitude data when the current position is input.

In actual block division, as in the example of Japan shown in FIG. 9, for example, the latitude and longitude are divided in units of 1 degree, the latitude division areas are A1 to A16, and the longitude division areas are B1 to B17. By doing so, the intersecting portion becomes a latitude / longitude division block. Due to such block division, for example, most of Tokyo has blocks (A6,
B11). Further, in such block division, the data may be appropriately subdivided or summarized according to the data amount of the place name and facility name included in each block.

Next, based on the current position belonging block obtained as described above, peripheral blocks within a predetermined range of the block are selected (step S24). This operation is performed by the peripheral block selection unit 12 of FIG. 1 based on the data selected by the current position belonging block selection unit 11 to select a block within a predetermined range around the selected current position block. .

In selecting the actual peripheral blocks, FIG.
0 (a) can be selected. In this example, in the case of division by latitude and longitude as shown in FIG. 9, for example, if the current position is located in Yamanashi prefecture, the block ( A6, B10) is selected as the current position belonging block, and blocks (A7, B) around it are selected.
9), (A6, B9), (A5, B9), (A7, B1
0), (A5, B10), (A7, B11), (A6
A total of eight blocks B11) and (A5, B11) are selected as peripheral blocks.

After that, with respect to the current position belonging block obtained in step S23 as described above and the peripheral blocks within a predetermined range of the current position belonging block obtained in step S24, the speech recognition dictionary corresponding to those blocks is set to the latitude. Select from the longitude division block voice recognition dictionary 13 and select as a block corresponding voice recognition dictionary group, and combine them to form the current position corresponding block group voice recognition dictionary 14 (step S26). After that, the process returns to step S21 again, and the same operation is repeated thereafter.

When the dictionary as described above is created, all of these data are actually collected and temporarily recorded in the storage unit, or the latitude / longitude division block voice recognition dictionary 13 is simply used.
Of the dictionaries existing in, the dictionary currently used as the voice recognition dictionary may be listed and recorded. In the case of only creating the data listed as described above, when searching the phoneme data corresponding to the input voice in the voice recognition processing, when the phonetic data corresponding to the input voice is searched, the list is stored in the latitude / longitude division block voice recognition dictionary 13. The voice recognition process is executed by selecting only a dictionary of existing blocks and performing a search.

On the other hand, if it is determined in step S22 that the voice recognition dictionary selection process to be performed is not the first process, that is, if the voice recognition dictionary has already been selected, is the current position belonging block changed? Determine whether or not. That is, when the current position changes as the vehicle moves, it is detected whether or not the current position is out of the range of the previously selected current position belonging block. As a result, when it is determined that the current position belonging block has not changed, the process returns to step S21 again, the current position data is continuously taken in, and the same operation is repeated.

If it is determined in step S25 that the current position affiliation block has changed, the process proceeds to step S23, the current position affiliation block is selected in the same manner as described above, and then, in step S24, predetermined blocks around the current position affiliation block are selected. The peripheral blocks are selected and the same operation is repeated. In this embodiment, the block selection process of step S23 and subsequent steps is performed only when it is determined in step S25 that the current position belonging block has changed, so that the process of step S23 and subsequent steps is always performed each time the current position data is fetched. This is done to reduce the processing load in the place name / facility name dictionary selection process.

As a result of the voice recognition dictionary selection processing by moving the current position as described above, for example, in the current position corresponding block group including the current position belonging block and peripheral blocks shown in FIG. When entering from Tokyo to Yamanashi prefecture, the current position affiliation block becomes a block (A6, B11) as shown in FIG. 7B, and the peripheral blocks move accordingly as shown. Further, when the current location enters from Saitama prefecture in Tokyo and enters the latitude division area a6 to A7, the current position belonging block moves to the block (A7, B11) as shown in FIG. The blocks also move as shown. In this way, the dictionary for performing the voice recognition process can be changed according to the movement of the current position.

When narrowing down a plurality of candidates in step S7 of FIG. 2, they can be processed in order according to the operation flow shown in FIG. 4, for example. That is, when the voice recognition dictionary search processing is performed in step S5 of FIG. 2 and as a result it is determined in step S6 that there is not only one search result, that is, there are a plurality of candidates, FIG.
In this embodiment, these plural candidates are displayed on the screen so that the user can confirm them (step S31). It should be noted that it is possible to set the process to proceed immediately to the next step S32 without performing such a screen display, or to set the display only when the number of candidates is within a predetermined number.

In step S32, when inputting which category the place name or facility name intended by the user belongs, a category list is displayed for the convenience of the input, and voice recognition by category is performed. The unit 21 searches for phoneme data in the category-specific dictionary 15 (step S33).

Whether or not there is only one candidate obtained as a result of the search is determined (step S34), and when it is determined that there is not only one candidate, that is, there are a plurality of candidates,
The distances from the current position are calculated for these plurality of candidates (step S35), rearranged in the order of decreasing distance, and displayed in a list (step S36).

As a result, for example, when the user wants to shop at 7-Eleven at the current position in FIG. 10C and designates 7-Eleven at a convenience store as a category, a list as shown in FIG. 11 is displayed. . By looking at such a list, the user can see that the nearest 7-11 Inabago store is located. Instead of the list as described above, a map centering on the present location may be displayed and the facilities as described above may be displayed on the map. The user selects a desired facility name or the like while looking at the above list display or map display (step S37), and outputs the selection result (step S).
38).

The voice recognition device for navigation according to the present invention operates as described above, but it is similarly used in countries other than the navigation device used in Japan as shown in FIGS. 9 and 10 above. In particular, in the United States, there are many place names and facility names in the vast land, and it is too heavy for the speech recognition processing device to use them in one speech recognition dictionary. The effect is particularly great when the present invention is used as a voice recognition device of a navigation device used in. At that time, for example, as shown in FIG. 12, the voice recognition dictionary is divided into latitude and longitude. In this example, the latitude division area n1
.About.n10 and longitude division regions w1 to w24 are used, and an area where these intersect is used as a block unit for partitioning the speech recognition dictionary. It should be noted that in the United States, in particular, since there are many administrative divisions of states along the latitude and longitude, the method of dividing blocks by latitude and longitude according to the present invention can be easily adapted.

Therefore, in the example shown in FIG.
When the current location exists in the block (n7, w15),
Eight peripheral blocks are selected in the same manner as the above-mentioned Japanese example, and the voice recognition processing is performed using the voice recognition dictionary of a total of nine blocks. As a result, the voice recognition dictionary is changed as the current position moves from FIG. 13 (a) as shown in FIGS. 13 (b) and 13 (c). The illustrated embodiment shows an example in which both latitude and longitude are set in 2.5 degree units, but for example, as shown in FIG. 14, it may be divided into fine blocks in 1 degree units. In that case, as shown in the same figure, the high-density population on the west coast of the United States can be divided into small blocks every one degree, and the other portions can be divided into blocks every two degrees. Further, a region with a low population density such as the central part of the inland US may be set to be divided into larger blocks.

Although the present invention can be implemented in various modes as described above, it is possible to select peripheral blocks in various modes by using a voice recognition dictionary divided by latitude and longitude in various modes. it can. Further, the narrowing-down processing when a plurality of candidates are present as a result of the voice recognition can also be narrowed down in another mode other than the above-described embodiment.

[0082]

Since the present invention is constructed as described above,
The navigation speech recognition apparatus according to the present invention, in order to solve the above problems, phoneme data of the place name and facility name included in the block divided by latitude and longitude and its related data,
A block unit voice recognition dictionary storage unit having a plurality of block unit voice recognition dictionaries recorded for each block, a current position belonging block to which the current position belongs, and a peripheral block in a predetermined range around it are selected. A current position corresponding block group selecting unit and a block unit voice recognition dictionary of the block group selected by the current position corresponding block group selecting unit,
A block position voice recognition dictionary corresponding to the current position selected from the block unit voice recognition dictionary storage unit is provided, and phoneme data corresponding to the voice of the input place name or facility name is searched from the voice block dictionary corresponding to the current position. Since a voice recognition processing unit for outputting is provided, when performing voice recognition processing using a voice recognition dictionary in which place names and facility names are set for each region, a suitable voice is not added to the voice recognition processing device and does not impose a heavy processing load. The recognition process can be performed, and the same process as the process of performing the voice recognition process from a large amount of data can be performed using a small voice recognition dictionary. As a result, it is possible to perform voice recognition accurately at high speed using an inexpensive data processing device.

Further, in another navigation voice recognition apparatus according to the present invention, the blocks divided by the latitude and longitude are so large that the place names and the facility names included in each block are as uniform as possible. Since it is divided by changing the size, it is possible to reduce the extent to which the voice recognition processing load changes according to the area where the current location is, and to perform voice recognition accurately at high speed using an inexpensive data processing device. It can be carried out.

Further, in another voice recognition device for navigation according to the present invention, since the blocks around the block to which the current position belongs are blocks adjacent to the current position belonging block, the area to be subjected to the voice recognition. When selecting, the selection can be easily performed.

In another navigation voice recognition apparatus according to the present invention, a block surrounding the block to which the current position belongs, a block adjacent to the block to which the current position belongs, and a block in a predetermined range around the block. Since it is included, it is possible to set the area targeted for voice recognition to an arbitrary range, it is possible to select and set the range as appropriate depending on the processing capacity of the voice recognition processing device, etc.
It is possible to appropriately select the range to be the target of voice recognition according to various situations.

In another navigation voice recognition apparatus according to the present invention, the current position of the block around the block to which the current position belongs is moved so that the amounts of the place name and the facility name are as equal as possible. Since there are many place names and facility names depending on the area where the current position exists, the total number of place names and facility names included in the block group for voice recognition can be changed. Can be equalized, the processing load of voice recognition is equalized,
Appropriate voice recognition processing can be performed as a whole.

Further, in another navigation voice recognition apparatus according to the present invention, the voice recognition processing unit exists in a block close to the current position for a place name or a facility name included in the current position corresponding block group voice recognition dictionary. Since the voice approximation degree is set to be higher, the facility intended by the user can be appropriately searched, for example, when searching for a nearby convenience store. In addition, since the degree of approximation is calculated in block units, the memory capacity may be smaller than that required for each place name or facility name in units of states, prefectures, etc., which makes arithmetic processing easier and speech recognition processing easier. The burden is reduced.

Further, in another navigation voice recognition apparatus according to the present invention, the current position corresponding block group voice recognition dictionary comprises a list in which blocks of the current position corresponding block group are recorded, and the voice recognition processing section, A place name or facility name having phoneme data corresponding to the input voice is searched from the dictionary data of the block recorded in the list of the block unit voice recognition dictionary storage unit, based on the list recording the block. Therefore, the data recorded in the current position-corresponding block group voice recognition dictionary need only be list data, and the data rewriting processing load can be reduced, so that an inexpensive voice recognition device can be obtained.

In another voice recognition device for navigation according to the present invention, the current position belonging block selection unit changes the block to which the current position belongs due to the movement of the current position and selects a new current position belonging block. At this time, since the surrounding block selection process is performed, it is not necessary to perform the surrounding block selection process each time the current position moves, and the voice recognition dictionary selection process can be simplified.

Further, another navigation voice recognition apparatus according to the present invention comprises a plurality of candidate narrowing-down processing units, and the plurality of candidate narrowing-down processing units performs voice recognition processing by the current position corresponding block group voice recognition dictionary. As a result, when a plurality of candidates are present, the narrowing down process is performed. Therefore, when a plurality of candidates are present as a result of performing the voice recognition process using the current position corresponding block group voice recognition dictionary, an appropriate one is selected. Only the items can be narrowed down and output, and the voice recognition device can be easily used.

Further, another navigation voice recognition device according to the present invention comprises a category-based voice recognition dictionary in which place names and facility names are recorded by category, and the plural candidate narrowing down processing section is a category designated by the user. Since the voice recognition processing is performed using the category-based voice recognition dictionary corresponding to, when there are a plurality of candidates as a result of the voice recognition processing using the current position corresponding block group voice recognition dictionary,
By using a separately prepared category-based voice recognition dictionary, it is possible to perform a search corresponding to the category designated by the user, and more accurate voice recognition processing can be performed.

Further, in another voice recognition device for navigation according to the present invention, the plurality of candidate narrowing down processing units calculate the distance from the current position for each of the plurality of candidates obtained as a result of the voice recognition processing, and the candidates are calculated. The output of the point distance order arrangement unit causes the display unit to display a list of a plurality of candidates in order, and the user can make a selection by this, so that the candidate closer to the current position among the plurality of candidates or other conditions are added. The user can arbitrarily select the closest one and can correctly recognize the place name or facility name intended by the user.

Further, another navigation voice recognition device according to the present invention calculates the distance from the current position for each of the plurality of candidates obtained as a result of the voice recognition processing, and outputs the candidate having the shortest distance as the voice recognition result. I did so,
Of the multiple candidates, the one that seems to be the most appropriate can be automatically output, which does not bother the user.
The voice recognition device can be easily used.

[Brief description of drawings]

FIG. 1 is a functional block diagram of an embodiment of the present invention.

FIG. 2 is a basic operation flow chart of the embodiment.

FIG. 3 is an operation flow diagram for performing a current location corresponding place name / facility name voice recognition dictionary setting process in the basic operation flow diagram.

FIG. 4 is an operation flow diagram for performing a process of narrowing down a plurality of candidates in the basic operation flow diagram.

FIG. 5 is a diagram showing an example of dividing a voice recognition dictionary according to the present invention by latitude and longitude and selecting a current position belonging block and a peripheral block; (a) shows an example of a latitude / longitude divided block; ) Shows an example of selecting the current position belonging block and peripheral blocks, (c) shows an example in which the current position moves between blocks, and (d) changes the voice recognition dictionary when the current position belonging block is changed. It is a figure which shows an example.

FIG. 6 is a diagram showing an example of selecting peripheral blocks in the present invention, FIG. 6A shows an example of setting peripheral blocks in a plurality of modes, and FIG. 6B is a diagram showing an example of setting in other modes. is there.

FIG. 7 is a diagram showing an example of subdividing a specific area when dividing the voice recognition dictionary according to the present invention by latitude and longitude,
(A) shows an example of division in one kind of mode, and (b) shows 2
It is a figure which shows the example divided into a mode of a kind.

FIG. 8 is a diagram showing an example in which the current position corresponding block group changes due to movement of the current position when subdividing a specific area when dividing the voice recognition dictionary according to the present invention by latitude and longitude.

FIG. 9 is a diagram showing an embodiment in which the method for creating a voice recognition dictionary that divides by latitude and longitude according to the present invention is applied to create a voice recognition dictionary for place names / facility names throughout Japan.

FIG. 10 is a diagram showing a current position corresponding block group for voice recognition in the embodiment of Japan.
FIG. 6C is a diagram showing an example in which the current position corresponding block group changes in accordance with the movement of the current position.

FIG. 11 is a diagram showing an example in which a plurality of candidates obtained as a result of the voice recognition process are narrowed down in a specific category, arranged according to the distance from the current position, and displayed on the screen.

FIG. 12 is a diagram showing an embodiment in which the method of creating a voice recognition dictionary for dividing by latitude and longitude according to the present invention is applied to the creation of a voice recognition dictionary of US place names / facility names.

FIG. 13 is a diagram showing a current position corresponding block group for voice recognition in the embodiment in the United States, and
FIG. 6C is a diagram showing an example in which the current position corresponding block group changes in accordance with the movement of the current position.

FIG. 14 is a diagram showing an example in which the voice recognition dictionary is subdivided with respect to the west coast portion where there are particularly many place names and facility names in the same embodiment.

FIG. 15 is a functional block diagram showing an example of a navigation speech recognition device to which the present invention is applied, which has been conventionally used.

[Explanation of symbols]

1 microphone 4 Speech recognition section 5 Speech recognition dictionary 6 place name, facility name dictionary section 7 Other dictionaries such as operating functions 9 Current position corresponding block group selection section 10 Current position input section 11 Current position affiliation block selection section 12 Peripheral block selection section 13 Latitude / longitude division block unit voice recognition dictionary storage unit 14 Current Position Corresponding Block Speech Recognition Dictionary 15 Speech recognition dictionary by place name / facility name category 20 Multiple candidate narrowing down processing unit 21 Voice recognition section by category 22 Candidate site distance order array narrowing down unit 23 Multiple candidate list display output unit 26 Output confirmation section

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G10L 15/22 G10L 3/00 571V 15/28 F term (reference) 2C032 HA21 HB06 2F029 AA02 AB01 AB07 AB13 AC02 AC18 5B075 ND14 NK37 PP07 PQ02 QP05 UU14 UU40 5D015 KK02 5H180 AA01 BB13 CC12 FF04 FF05 FF22 FF25 FF33

Claims (12)

[Claims] 1. A block unit voice recognition provided with a plurality of block unit voice recognition dictionaries in which phoneme data of place names and facility names included in blocks divided by latitude and longitude and related data are recorded for each block. A dictionary storage unit, a current position belonging block to which the current position belongs, and a peripheral block in a predetermined range around it, and a block selected by the current position corresponding block group selecting unit and the current position corresponding block group selecting unit A block unit voice recognition dictionary of a group, and a current position corresponding block group voice recognition dictionary selected from the block unit voice recognition dictionary storage unit, phoneme data corresponding to the voice of the input place name or facility name, the current position A voice recognition device for navigation, comprising a voice recognition processing unit for searching and outputting from a corresponding block group voice recognition dictionary. 2. The block divided by the latitude and the longitude is
2. The navigation voice recognition device according to claim 1, wherein the block size is divided so that the amounts of the place name and the facility name included in each block are as equal as possible. 3. The voice recognition device for navigation according to claim 1, wherein the blocks around the block to which the current position belongs are blocks adjacent to the block to which the current position belongs. 4. The navigation system according to claim 1, wherein the blocks around the block to which the current position belongs include blocks adjacent to the block to which the current position belongs and blocks in a predetermined range around the block. Speech recognizer. 5. The block around the block to which the current position belongs is changed along with the movement of the current position so that the amounts of the place name and the facility name are as equal as possible. Speech recognition device for navigation. 6. The voice recognition processing unit is configured to set a voice approximation degree higher for a place name or a facility name included in the current position corresponding block group voice recognition dictionary, as the block is closer to the current position. The voice recognition device for navigation according to claim 1. 7. The current position corresponding block group voice recognition dictionary comprises a list in which blocks of the current position corresponding block group are recorded, and the voice recognition processing unit includes a place name or a place name having phoneme data corresponding to the inputted voice. 2. The navigation voice recognition device according to claim 1, wherein the facility name is searched from the dictionary data of the block recorded in the list of the block unit voice recognition dictionary storage unit based on the list in which the block is recorded. . 8. The current position affiliation block selection unit selects a new current position affiliation block when the block to which the current position belongs is changed due to the movement of the current position,
The voice recognition device for navigation according to claim 1, wherein a peripheral block selection process is performed. 9. A plural candidate narrowing down processing unit, wherein the plural candidate narrowing down processing unit performs narrowing down processing when there are a plurality of candidates as a result of the speech recognition processing by the current position corresponding block group speech recognition dictionary. The voice recognition device for navigation according to claim 1, which is performed. 10. A category-based voice recognition dictionary in which place names and facility names are recorded according to categories, wherein the plural candidate narrowing down processing unit uses the category-based voice recognition dictionary corresponding to a category designated by a user for voice recognition. The navigation voice recognition device according to claim 9, wherein recognition processing is performed. 11. The multiple candidate narrowing-down processing unit includes a candidate point distance order arrangement unit that calculates a distance from the current position for each of the plurality of candidates obtained as a result of the voice recognition process and arranges the distances in order of distance according to the calculation result. 10. The navigation voice recognition device according to claim 9, wherein a plurality of candidates are sequentially displayed in a list on the display unit by the output of the candidate point distance order arrangement unit, and the user makes a selection accordingly. 12. The navigation voice according to claim 9, wherein a distance from the current position is calculated for each of the plurality of candidates obtained as a result of the voice recognition processing, and the candidate having the shortest distance is output as the voice recognition result. Recognition device.