JP4869642B2 - Voice recognition apparatus and vehicular travel guidance apparatus including the same - Google Patents

Description

  The present invention relates to a speech recognition device and a vehicle travel guidance device including the same.

  2. Description of the Related Art In recent years, a vehicular travel guidance device that receives position information where a vehicle is currently traveling from an artificial satellite, displays vehicle position information together with map information on a display unit, and guides a driver has been put into practical use. Further, a vehicle travel guidance device that has a voice recognition device for enabling operations such as destination setting by voice has been put into practical use. This speech recognition device is intended to facilitate the operations required by the user. However, the number of target words for speech recognition is enormous, such as place names and facility names, and speech recognition processing based on comparison with input speech inevitably takes time, and the possibility of misrecognition increases. ing.

For this reason, Patent Document 1 is equipped with a speech recognition device that facilitates operations such as destination setting, shortens the speech recognition processing time, and reduces the possibility of erroneous recognition, and the speech recognition device. A navigation device has been proposed. According to the proposed technique of Patent Literature 1, it is possible to narrow down processing target data to be subjected to speech recognition processing, for example, to a designated map range from comparison data stored in a recognition dictionary. As a result, the voice recognition processing time can be shortened and the possibility of erroneous recognition can be reduced. As a result, the operation of selecting a desired target from the recognition results of the input voice can be facilitated. is there.
JP 2004-20883 A

  However, the speech recognition apparatus has the following problems in addition to the problems described above. The speech recognition dictionary provided in the speech recognition apparatus has comparison data as a comparison target for recognizing input speech. Of the comparison data, comparison data related to place names (hereinafter referred to as place name comparison data) is generally organized in units of regions, and the entire place name comparison data is organized as a collection of items organized in units of regions. Is done. When the place name comparison data as a whole is configured in this way, if place names with the same name exist in a plurality of regions, place name comparison data corresponding to each of them exists. For example, if the entire place name comparison data covers the United States and all over Japan, and the region is a state or prefecture, the place name comparison data exists for each place name with the same name. Data. In addition, since the place name comparison data is a comparison target when recognizing the input voice, the voice recognition processing time increases as the number of place name comparison data increases. In addition, as a result of recognition of the input speech, since a plurality of place names with the same name existing in different regions are listed as a plurality of candidate place names, there is a problem that it takes time for the user to find and select a desired destination from among the recognition results. Will occur.

  The present invention has been made in view of the above problems, and suitably suppresses the volume of comparison data included in the speech recognition dictionary, and at the same time, shortens the speech recognition processing time and the user desires from the recognition results of the input speech. It is an object of the present invention to provide a voice recognition device capable of improving the operability of an operation for selecting a destination and a vehicle travel guidance device including the same.

In order to solve the above problem, the present invention compares the speech recognition dictionary storing comparison data for recognizing input speech with the input speech data generated based on the input speech and the comparison data. A speech recognition apparatus comprising: a speech recognition unit that recognizes an input speech and outputs a recognition result of the input speech, wherein the speech recognition dictionary includes place name comparison data having place names as display information; comparison data is a unique structure in place names unit, further, different regions in the viewing including the area number data to display information number local place names of the same name exists, and when the input speech is a place name, the The region comparison data satisfying a predetermined matching degree with the recognition result of the input speech is selected as a candidate place name, and the candidate place number which is the number of area data corresponding to the candidate place name is selected together with the candidate place name. Characterized in that it comprises a display information generation unit for generating a first display information for Shimesuru. According to the present invention, by configuring the entire place name comparison data by place name, in other words, by creating a speech recognition dictionary as a unique dictionary by place name, place names having place names having other duplicate names as data contents Comparison data can be excluded from the speech recognition dictionary. Thereby, it is possible to suppress the capacity of place name comparison data included in the speech recognition dictionary. Moreover, it is possible to shorten the time required for the voice recognition processing by suppressing the capacity of the place name comparison data. Furthermore, since a place name having the same name does not appear as a plurality of candidates as a recognition result of the input voice, it is possible to improve the operability of an operation in which the user selects a desired destination from the recognition result. Note that even if the place names are the same, place names represented by different character strings when represented by the same kind of character string such as kanji or alphabet are not treated as place names of the same name.
Further, according to the present invention, it is possible to realize display for notifying the user whether or not a place name with the same name exists in a plurality of areas in various modes based on the place name comparison data. Note that the speech recognition dictionary may have other data associated with the place name comparison data, and the other data may have the number of areas where the place names exist based on the place name comparison data as display information.
And according to this invention, the 1st display information for notifying a user whether the candidate place name exists in a some area with the candidate place name based on a recognition result is realizable with various display modes.

  In addition, the present invention includes a control unit and a reading mechanism that reads map data from a map database having road maps, facility information, and the like as data contents. May select the map data having the place name including the selected candidate place name as data contents from the map database. Here, unlike the place name comparison data, the map data has a place where the place name exists together with the place name as display information. According to the present invention, even if the selected candidate place names exist in a plurality of areas, it is possible to specify each place name having the same name.

  Further, according to the present invention, the display information generation unit displays second display information for displaying the name of the area where the place name exists together with the place name based on the map data searched from the map database by the control unit. It may be generated. According to the present invention, the second display information for informing the user of a plurality of place names having the same name together with the area where the place names exist can be realized in various display modes. Thus, the user can recognize and select in which region the desired destination exists.

  In the present invention, the display form of the place names of the first display information may be a list format in the order of the matching degree of the recognition result. According to the present invention, it becomes easy for the user to find and select a desired destination from among the recognition results.

  In the present invention, the display mode of the place name of the second display information may be an alphabetical list form of the area name. According to the present invention, the user can easily find and select a desired destination by using the initial of the area name as a clue.

Further, the present invention is the vehicle travel guidance device including the voice recognition device, wherein the display mode of the place name of the second display information is a list format in order from the current position of the vehicle. May be. According to the present invention, when a user inputs by voice a place name that is not so far from the current position of the vehicle, the user can easily select a desired destination from the display based on the second display information. It is possible to search and select.

  According to the present invention, it is possible to suitably suppress the volume of comparison data included in the voice recognition dictionary, simultaneously reduce the voice recognition processing time, and operate the user to select a desired destination from the input voice recognition results. It is possible to provide a voice recognition device capable of improving the performance and a vehicle travel guidance device including the same.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.

  A configuration of a voice recognition device 1A and a navigation device 10A including the voice recognition device 1A according to the present embodiment will be described in detail with reference to FIG. In FIG. 1, the configuration indicated by the double line box is the configuration of the speech recognition apparatus 1 </ b> A according to the present embodiment. In FIG. 1, a reading mechanism 11 reads a recording medium (map database) such as a CD (Compact Disc) -ROM (Read Only Memory) or a DVD (Digital Versatile Disc) -ROM storing map data and other guidance data. It is the structure for. However, the present invention is not limited to this, and a storage device such as a hard disk drive can also be applied.

  The operation unit 12 is configured for a user to input various operations and settings to the in-vehicle navigation device 10A. This may be configured individually such as a remote controller device or a control panel, or may be configured as a touch panel type input device formed integrally with the display unit 16 described later. Furthermore, it may be constituted by a microphone for voice input.

  The VICS (road traffic information communication system) receiving unit 13 is configured to receive VICS information transmitted from a radio wave beacon or an optical beacon. The GPS (Global Positioning System) receiving unit 14 is configured to receive a GPS signal transmitted from a GPS satellite and detect the latitude and longitude of the current position of the vehicle. The self-contained navigation sensor 15 includes an angle sensor 15a including a gyro for detecting the vehicle direction, and a distance sensor 15b that generates a pulse for every predetermined travel distance. Detect speed.

  The display unit 16 is configured by, for example, a liquid crystal display type display device or the like, and is configured to display various information such as a map, a guide route, a current position of a vehicle, a building, and other icons input from the navigation device 10A. is there. The speaker 17 is similarly configured to output voice guidance information or the like input from the navigation device 10A. In addition, the speaker 17 can also output music input from an audio device or the like.

  In the navigation device 10A, the buffer memory 21 is configured to temporarily store map data or the like input from the reading mechanism 11 under the control of the control unit 27 described later. The I / Fs (interfaces) 22, 23, 24, and 25 are configured to connect the operation unit 12, the VICS receiving unit 13, the GPS receiving unit 14, the self-contained navigation sensor 15, and the internal bus in the in-vehicle navigation device 10A, respectively. It is.

  The control unit 27 includes an arithmetic processing device such as a microcomputer or a CPU (Central Processing Unit). The control unit 27 has a built-in navigation program, and a map to be detected or displayed based on signals output from the GPS receiving unit 14 and the autonomous navigation sensor 15 according to the program. And the like are read from the reading mechanism 11 to the buffer memory 21, the guide route is searched from the map data read to the buffer memory 21, and the guide route that matches the set search condition is read to the buffer memory 21. Various processes related to navigation such as searching for one or more using data or the like are executed. However, for example, the above program can be stored in a CD-ROM, a DVD-ROM, or the like. In this case, the control unit 27 reads and executes it as necessary.

  The map drawing unit 28 is configured to perform a map image drawing process using the map data read to the buffer memory 21. The display information generation unit 29 is configured to generate various menu screens (operation screens) and various marks such as a cursor according to the operation status, and includes first display information and second display information described later. It is the structure for producing | generating. The guide route storage unit 30 stores data on all nodes of the guide route searched for in the control unit 27 based on the starting point, destination and other search conditions, and data of guide routes changed during the search (hereinafter referred to as these Is referred to as guidance route data). The guide route drawing unit 31 is a configuration for reading the guide route data from the guide route storage unit 30 and drawing the guide route in a display mode (highlighted display using colors and line widths) different from other roads. . The icon drawing unit 32 has a configuration for drawing an icon such as a building or a spot to be drawn on a map image, a vehicle, or another vehicle. The audio output unit 33 includes, for example, a DSP (Digital Signal Processor) and the like, and outputs an audio signal to the speaker 17 based on a signal from the control unit 27.

  The image composition unit 34 includes a guide route drawn by the guide route drawing unit 31, an operation screen and various marks drawn by the display information generation unit 29, an icon drawing unit, and a map image drawn by the map drawing unit 28. In this configuration, various icons drawn in 32, images input from the control unit 27, and the like are appropriately overlapped and displayed on the display unit 16.

  Next, the configuration of the speech recognition apparatus 1A according to the present embodiment shown in FIG. 1 will be described in detail. As described above, the configuration of the speech recognition apparatus 1A is a configuration indicated by a double box in FIG. When the navigation device 10A includes the voice recognition device 1A, the configuration of the control unit 27, the display information generation unit 29, and the like can be shared with the navigation device 10A as shown in FIG. That is, the control unit 27 and the display information generation unit 29 are the configuration of the navigation device 10A and the configuration of the voice recognition device 1A.

  The microphone 2 is configured to convert voice input by the user into an electrical signal. However, in the case where the operation unit 12 of the navigation device 10 </ b> A is configured by a voice input microphone, the operation unit 12 can be used without providing the microphone 2 separately. The voice recognition dictionary 3A is composed of a recording medium such as a CD-ROM or DVD-ROM that stores comparison data to be compared for recognizing input voice. However, it is also possible to apply a storage device such as a hard disk drive, and when the map database is composed of a storage device storing map data, it is stored in a mode that uses a part of the capacity of this storage device. It is also possible to share devices. The voice recognition unit 4 has a configuration for recognizing an input voice by further generating an input voice converted into an electric signal into input voice data of a digital signal and comparing it with comparison data. However, the microphone 2 may generate digital signal input voice data, and an A / D converter may be provided between the microphone 2 and the voice recognition dictionary 4.

  Next, comparison data stored in the speech recognition dictionary 3 will be described in detail. Of the comparison data, comparison data related to place names (hereinafter referred to as place name comparison data) is generally organized in units of regions, and the entire place name comparison data is organized as a collection of items organized in units of regions. Is done. The whole place name comparison data is configured in this way because the target range for creating comparison data covers a wide range, for example, all over Japan and the United States, so it is more convenient to create place name comparison data for each region first. by.

  When the entire place name comparison data is configured in such a region unit, if there are place names with the same name in a plurality of areas, there are a plurality of place name comparison data having place names with the same name corresponding to each area as data contents. Become. Here, for convenience of explanation, the speech recognition dictionary 3 having place name comparison data configured in units of regions is referred to as a speech recognition dictionary 3B, and the speech recognition device 1 including the speech recognition dictionary 3B is referred to as a speech recognition device 1B. In addition, the navigation device 10 including the voice recognition device 1B is referred to as a navigation device 10B.

  FIG. 2 is a diagram schematically showing a part of the place name comparison data configured in units of regions. As shown in FIG. 2, the place name comparison data more specifically includes reading data for comparison with the input voice data and character string data for displaying the place name as a character string on the display unit 16. Yes. That is, the character string data has the place name as display information. Here, “KIMBERLY” is the name (place name) of a city existing in a plurality of states (regions) in the United States. In such a case, the place name comparison data configured in units of area includes place names such as “KIMBERLY, AL”, “KIMBERLY, ID” and the place names together with the reading data having “Kimberley” as the data content as shown in FIG. Has character string data whose data content is a character string composed of the region where the character exists. “AL” and “ID” are abbreviations of states. The reason why the same reading data is set in the place name comparison data having the place names of the same names existing in other areas as the data contents is to avoid the necessity of specifying the place names in a complicated manner as input speech. .

  FIG. 3 is a diagram illustrating a method for searching for a destination by using input speech in the navigation device 10B including the speech recognition device 1B. As shown in FIG. 3, when the user searches for a destination by voice, the navigation device 10B displays the display screen 41 on the display unit 16 and inputs the name of the city (place name) by voice to the user. Prompt. The display screen 41 is displayed when the user selects the voice input mode when searching for the destination. Subsequently, the user inputs a voice “Kimberly” into the microphone 2. In response to the input voice “Kimberley”, the navigation device 10B displays the candidate place names of the recognition result of the voice recognition device 1B in a list format as shown on the display screen 42. On the other hand, the user inputs “list number four” into the microphone 2 by voice. In response to the input voice “list number four”, the navigation device 10B displays the road map of the candidate place name “KIMBERLY, MO” listed fourth in the list as a destination in a display mode as shown on the display screen 43.

  As shown in the display screen 42 of FIG. 3, when the place name comparison data is configured in units of regions, the speech recognition unit 4 performs “KIMBERLY, AL” “KIMBERLY, ID” for the input speech “Kimberley”. Are recognized as candidate place names. Further, since the voice recognition unit 4 determines the identity by comparing the input voice data and the place name comparison data by matching, when the predetermined matching degree is satisfied, for example, “KIMBERLINGCITY, MO”, “KIMBERTON, PA”, etc. Is also listed as a candidate place name. As described above, when the place name comparison data is configured in units of areas, place names satisfying a predetermined matching degree are listed as candidate place names in addition to place names having the same name existing in different areas. Therefore, it takes time for the user to find and select a desired destination from the recognition results as shown on the display screen 42.

  In response to the above-described problem, the speech recognition dictionary 3A according to the present embodiment stores place name comparison data that makes the place name unique. FIG. 4 is a diagram schematically showing a part of place name comparison data stored in the speech recognition dictionary 3A. As shown in FIG. 4, in the speech recognition dictionary 3A, for the reading data having “Kimberley” as the data content, the place name comparison having the character string data having the character string “KIMBERLY” as the data content There is only data. That is, in the speech recognition dictionary 3A, a character string such as “KIMBERLY, AL” or “KIMBERLY, ID” is added to the read data having “Kimberly” as the data content like the speech recognition dictionary 3B shown in FIG. There is no place name comparison data having character string data as data contents. Furthermore, the place name comparison data stored in the speech recognition dictionary 3A has area number data that uses the number of areas in which the place names exist as display information, unlike the place name comparison data stored in the voice recognition dictionary 3B.

  Next, a method for searching for a destination using input voice in the navigation apparatus 10A including the voice recognition apparatus 1A will be described in detail with reference to FIGS. FIG. 5 is a diagram illustrating a method for searching for a destination by using input voice in the navigation apparatus 10A including the voice recognition apparatus 1A. FIG. 6 is a diagram schematically showing a process in which the speech recognition dictionary 3A and the map database are used in generating the display screens 52 and 53 shown in FIG. As shown in FIG. 5, when the user searches for a destination by voice, the navigation device 10A displays the display screen 51 on the display unit 16 and inputs the name of the city (place name) by voice to the user. Prompt. On the other hand, the user inputs “Kimberly” into the microphone 2 by voice. In response to the input voice “Kimberley”, the navigation device 10A displays the candidate place names of the recognition result of the voice recognition device 1A in a list format as shown in the display screen 52.

  In the process of generating the display screen 52, the speech recognition dictionary 3A is used as shown in FIG. More specifically, the voice recognition unit 4 converts the input voice “KIMBERLY” into input voice data, and recognizes the input voice “KIMBERLY” by comparing with the place name comparison data of the voice recognition dictionary 3A (arrow 1). Subsequently, the voice recognition unit 4 selects place name comparison data that satisfies the input voice “KIMBERLY” and a predetermined matching degree. Here, the place name that the selected place name comparison data has as data contents becomes the candidate place name. When the place name comparison data having a candidate place name is selected, the display information generation unit 29 generates first display information for displaying the candidate place name and the number of candidate areas corresponding to the candidate place name. The first display information is generated by reading place name comparison data having candidate place names selected based on instructions from the control unit 27 as data contents from the speech recognition dictionary 3A to the display information generating unit 29. When the first display information is generated, the control unit 27 displays the first display information on the display unit 16 in a display mode as shown on the display screen 52 (arrow 2). The display screen 52 is provided with a region number display 52a for displaying the number of regions where the candidate place name on which the cursor 55 is placed. The number of areas displayed in the area number display 52a is displayed based on the place name comparison data having the candidate place name where the cursor is placed as the data content. That is, each time the cursor 55 moves to another candidate place name, area number data is read from the place name comparison data having the candidate place name as data content to the display information generating unit 29 based on the command of the control unit 27, and the display information is displayed. The generation unit 29 changes the region number display 52a based on the read region number data. However, the display information generation unit 29 may generate the first display information in a display mode that displays the number of areas for each candidate place name. As shown in the display screen 52, the user can know that “KIBALL” where the cursor 55 is hit exists in 15 states in the United States by this area number display 52 a.

  As shown in the display screen 52, “KIMBERLY”, “KIMBALL”, “KINBRAE”, “KINGSDALE”, “KINGSTON”, “KINNEY”, and the like are listed as the candidate place names for the input voice “Kimberly”. As described above, even if the input voice is “Kimberley”, other place names can be listed as candidate place names if the predetermined matching degree is satisfied. However, in this embodiment, since the voice recognition dictionary 3A is created as a unique dictionary with place names, place names with the same name existing in other areas are not listed as a plurality of candidate place names, and the user can easily “KIMBERLY”. It is possible to find out.

  The first display information can be displayed in the following display mode in addition to the display mode as shown on the display screen 52. As described above, the place name comparison data having candidate place names displayed on the display screen 52 is selected by the voice recognition unit 4 according to the degree of matching with the input voice. Therefore, if the candidate place names of the first display information are displayed in a list form in descending order of the matching degree based on the accuracy of the input voice, it can be expected that the user can easily find a desired destination.

  As illustrated in FIG. 5, when the user selects “KIMBERLY” on the display screen 52, the navigation device 10 </ b> A displays the display screen 53. On the display screen 53, the place name “KIMBERLY” and area names such as “AL” and “ID” where the place name exists are displayed in a list format. In the process of generating the display screen 53, a map database is used as shown in FIG. More specifically, the control unit 27 acquires character string data of place name comparison data having the selected candidate place name “KIMBERLY” as data contents from the speech recognition dictionary 3A. Subsequently, the control unit 27 reads the map data from the map database to the reading mechanism 11 and further reads the map data to the buffer memory 21. The control unit 27 determines whether or not the character string “KIMBERLY” is included in the data content of the map data read to the buffer memory 21 based on the character string data. When the determination is completed, the control unit 27 again reads other map data from the map database to the reading mechanism 11 and further reads the map data to the buffer memory 21 until all the map data is determined (arrow 3). ). As described above, map data including the character string “KIMBERLY” as data contents can be searched from the map database.

  Subsequently, based on the map data retrieved from the map database, the display information generation unit 29 displays second display information for displaying the place name “KIMBERLY” and the area names “AL”, “ID” and the like where the place names exist. Generate. The second display information is generated by reading the map data retrieved from the buffer memory 21 to the display information generation unit 29 based on a command from the control unit 27. As shown in FIG. 6, the map data has display information composed of place names and area names where place names exist as data contents. When the second display information is generated, the control unit 27 displays the second display information on the display unit 16 in a display mode as shown on the display screen 53 (arrow 4). In this way, the display information generating unit 29 displays a plurality of place names “KIMBERLY” having the same name together with the areas “AL” and “ID” in which “KIMBERLY” is present on the display screen 53 as shown in FIG. Second display information can be generated.

  Further, the second display information can be displayed in the following display mode in addition to the display mode as shown on the display screen 53. On the display screen 53, the place names are displayed in a list format in alphabetical order of the area names. However, the place names inputted by the user for searching for the destinations often exist near the current position of the vehicle. . In consideration of such circumstances, it is effective for the user to display the place names of the second display information in a list format in the order closer to the current position of the vehicle. Here, since the place name of the second display information is based on the map data, it is possible to simultaneously grasp the location of the place name from this map data. Further, the current position of the vehicle can be detected by a GPS signal received by the GPS receiver 14. Based on these pieces of information, it is possible to recognize how far the place name of the second display information is from the current position of the vehicle. Accordingly, it can be expected that the user can easily find a desired destination from the display screen based on the second display information.

  When the user searches for and selects a desired destination from the display screen 53, the navigation device 10A displays a road map as the destination, like the display screen 54 (arrow 5). As a result, the user can know the desired destination. When the user selects a candidate place name that exists in only one area on the display screen 52, the display screen 54 is immediately displayed as shown in FIG.

  In addition, as a mode for displaying the first display information, the second display information, and the target road map, the following display modes are possible in addition to the display mode shown in FIG. For example, the first display information and the second display information are simultaneously displayed on one screen, or the target road map and the first display information or the second display information are simultaneously displayed on one screen. Is also possible. As an advantage of displaying on one screen at the same time, for example, when the cursor 55 for designating a candidate place name based on the first display information moves, a plurality of place names with the same name based on the second display information are displayed on the same screen. Therefore, there are advantages such as being easily transmitted to the user intuitively. Moreover, as a display mode of the first display information, it is possible to display in alphabetical order or display in descending order of matching as described above, but these display modes can be further selected by the user. It is also possible to do. In the present embodiment, the speech recognition dictionary 3A is created for the entire United States. For example, when targeting all over Japan, the candidate place names are displayed in the order of 50 tones as the display mode of the first display information. It is also possible. The modification of the display mode of the first display information can be similarly applied to the display mode of the second display information.

  Moreover, the following specific effects are mentioned as an effect of creating a speech recognition dictionary 3A that is unique by place name. When creating a speech recognition dictionary for the United States, if the place name comparison data is created for each state (region) for the city name (place name) and the entire place name comparison data is configured, the speech recognition dictionary contains 95,967. There will be place name comparison data. When this is uniquely created with a city name like the speech recognition dictionary 3A according to the present embodiment, the number of place name comparison data possessed by the speech recognition dictionary 3A is 60,350. Thus, in the speech recognition dictionary 3A according to the present embodiment, it is possible to suppress the data capacity. In addition, there are 48,641 place names that exist only in one state in the United States, and 11,709 place names exist in a plurality of states. Furthermore, the city name “FRANKLIN”, which exists in most states, exists in 35 states. When a plurality of these place names are displayed as candidate place names like the display screen 41 shown in FIG. 3, it can be easily imagined that it takes a long time for the user to select a desired destination. Based on the speech recognition dictionary 3A according to the present embodiment, for example, only one “FRANKLIN” is displayed as in the display screen 52 shown in FIG. 4, so the user can easily find a desired destination from the candidate place names. Can be selected. As described above, the volume of the comparison data included in the voice recognition dictionary 3 is suitably suppressed, and at the same time, the voice recognition processing time is shortened and the operability of the user selecting a desired destination from the input voice recognition results is improved. It is possible to realize a voice recognition device 1A capable of achieving the above and a navigation device 10A including the same.

  The embodiment described above is a preferred embodiment of the present invention. However, the present invention is not limited to this, and various modifications can be made without departing from the scope of the present invention.

It is a figure which shows the structure of 10 A of navigation apparatuses provided with the speech recognition apparatus 1A and the speech recognition apparatus 1A which concern on a present Example. It is a figure which shows typically a part of place name comparison data comprised by the area unit. It is a figure which shows the method of searching a destination with input audio | voice in navigation apparatus 10B provided with the speech recognition apparatus 1B. It is a figure which shows typically a part of place name comparison data which the speech recognition dictionary 3A stores. It is a figure which shows the method of searching a destination with input audio | voice in navigation apparatus 10A provided with the speech recognition apparatus 1A. It is a figure which shows typically the process in which the speech recognition dictionary 3A and a map database are utilized in producing | generating each of the display screens 52 and 53 shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Voice recognition apparatus 2 Microphone 3 Voice recognition dictionary 4 Voice recognition part 10 Navigation apparatus 11 Reading mechanism 12 Operation part 13 VICS receiving part 14 GPS receiving part 15 Self-contained navigation sensor 15a Angle sensor 15b Distance sensor 16 Display part 17 Speaker 21 Buffer memory 22 , 23, 24, 25 I / F
27 Control unit 28 Map drawing unit 29 Display information generation unit 30 Guide route storage unit 31 Guide route drawing unit 32 Icon drawing unit 33 Audio output unit 34 Image composition unit 41, 42, 43 Display screen 51, 52, 53 by navigation device 10B 54 Display screen by navigation device 10A 52a Number of areas display 55 Cursor

Claims (6)

A speech recognition dictionary for storing comparison data for recognizing input speech, and comparing the input speech data generated based on the input speech with the comparison data to recognize the input speech, and the recognition result of the input speech A voice recognition device including a voice recognition unit for outputting
The voice recognition dictionary has place name comparison data having place names as display information,
The place name comparison data is a unique structure in place names unit further seen including local number of data to be the same name of the place name display information number area existing in the different regions,
When the input voice is a place name, the place name comparison data satisfying a predetermined matching degree with the recognition result of the input voice is selected as a candidate place name, and the area number data corresponding to the candidate place name together with the candidate place name. A speech recognition apparatus comprising: a display information generation unit that generates first display information for displaying the number of candidate areas . A control unit, and a reading mechanism for reading map data from a map database having data such as road maps and facility information,
When the user selects a candidate place name having a plurality of candidate areas from among the recognition results, the control unit includes map data having a place name including the selected candidate place name as data content in the map database. The speech recognition apparatus according to claim 1 , wherein search is performed from The display information generation unit generates second display information for displaying the name of the area where the place name exists together with the place name, based on the map data searched from the map database by the control unit. The speech recognition apparatus according to claim 2 . The speech recognition apparatus according to any one of claims 1 to 3 , wherein a display form of the place name of the first display information is a list format in descending order of matching degree of the recognition results. The speech recognition apparatus according to claim 3 or 4 , wherein a display form of the place name of the second display information is a list form of the area name in alphabetical order. 5. The travel guidance device for a vehicle provided with the voice recognition device according to claim 3 , wherein a display mode of the place name of the second display information is a list format in an order close to a current position of the vehicle.

Images