JP2010038751A - Navigation system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a navigation system with improved certainty in searching place names and road names by speech recognition. <P>SOLUTION: A speech recognition dictionary memory 120 memorizes place name speech recognition dictionary information and road name speech recognition dictionary information. When a user select a character or a character string by touching a touch panel 90 with fingers etc. under a condition that a character selection screen is displayed in a display 70, a controlling section 102 extracts character strings for place name and road name having in the forefront the character or the character string selected by the user, from among the place name speech recognition dictionary information and the road name speech recognition dictionary information memorized in the speech recognition dictionary memory 120. Subsequently, when the user speaks, a speech recognizing section 122 generates a character string for user's voice corresponding to the voice, and the controlling section 102 identify one most similar to the character string for user's voice from among the extracted character strings for place name and road name. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a navigation device that searches for a route to a destination set by a user and provides guidance to the destination.

  The navigation device searches for a route (guidance route) to a destination set by the user, and performs guidance by an image or sound so that the vehicle travels to the destination along the guidance route. Some of such navigation apparatuses have a function of recognizing a user's voice (see, for example, Patent Documents 1 and 2).

A conventional navigation device prepares a speech recognition dictionary including a character string of a place name and a road name that can be a destination. Then, the navigation device identifies a character string corresponding to the voice by voice recognition processing when the user utters a place name or a road name, and among the character strings of the place name and the road name included in the voice recognition dictionary, It has a function of specifying the most similar character string corresponding to the recognized user's voice as a destination candidate. The conventional navigation device extracts a keyword from a user's voice, and acquires appropriate map information based on the keyword and the user's interests.
JP 2004-20883 A JP-A-8-50698

  However, the above-described speech recognition dictionary needs to be provided for every place name or road name, which is a huge amount. For this reason, there is a problem that an error is highly likely to occur when a character string of a place name or a road name that is most similar to the character string corresponding to the user's voice is specified. On the other hand, the user utters a predetermined number of characters at the beginning of the place name or road name, narrows down the speech recognition dictionary to be searched based on the speech recognition result, and then the user selects the place name or road name. If the voice when the voice is uttered is recognized, the character string of the place name or road name included in the voice recognition dictionary after narrowing down is the most similar to the character string corresponding to the recognized user's voice. A method for identifying the same can also be considered. However, with this method, if an error occurs in speech recognition when a user utters a predetermined number of characters at the beginning of a place name or road name, a result unintended by the user is obtained.

  SUMMARY OF THE INVENTION An object of the present invention is to solve the above-described problems, and to provide a navigation device that improves the certainty of searching for place names and road names based on voice recognition.

  The navigation device according to the present invention includes a storage unit that stores character strings of place names and road names that can be set points, an acquisition unit that acquires a character or a character string selected by an operation of an operation unit by a user, and the storage unit Recognizing the character string extraction means for extracting the character string or the character string acquired by the acquisition means among the character strings of the place name and road name stored in the means, and the voice uttered by the user, A speech recognition unit that generates a recognition result, and a character string of a place name or a road name corresponding to the recognition result generated by the speech recognition unit among the place name and road name character strings extracted by the character string extraction unit. And a character string search means for searching.

  According to this configuration, when the user operates the operation unit to select a character or a character string, the character or character string selected by the user from among the character strings of the place name and the road name stored in advance is set to the top. When a user utters a speech, the speech recognition result is generated, and the place name or road name corresponding to the speech recognition result is extracted from the extracted place name and road name character strings. Searched. Therefore, since the character strings of the place names and road names stored in advance are narrowed down and searched based on the voice uttered by the user, the amount of place names or road names to be searched is reduced, and the search is performed. Certainty is improved. In addition, narrowing down of character strings of place names and road names stored in advance is performed based on characters or character strings selected by the user by operating the operation unit. Is not used, errors are unlikely to occur, and the reliability of search is improved.

  From the same point of view, in the navigation device according to the present invention, the voice recognition unit generates a character string corresponding to the voice uttered by the user as a recognition result, and the character string search unit uses the character string extraction unit. Of the extracted character strings of place names and road names, the character strings of the place names and road names that are most similar to the character strings corresponding to the voices uttered by the user generated by the voice recognition means are specified.

  The navigation apparatus according to the present invention further includes an output unit that outputs a character string of a place name or a road name searched by the character string search unit.

  In the navigation device according to the present invention, the output means outputs a voice corresponding to the character string of the place name or road name searched by the character string search means.

  The navigation device according to the present invention further includes display processing means for displaying a character selection screen on a touch panel type display unit, and the acquisition means selects a character selected by an operation of touching the display unit by the user. Or get a string.

  According to the present invention, since the place name and road name character strings stored in advance are narrowed down, the place name or road name corresponding to the voice uttered by the user is searched. Reduces the amount of road names and improves search certainty. In addition, since voice recognition is not used when narrowing down character strings of place names and road names stored in advance, errors are unlikely to occur and the reliability of search is improved.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings. FIG. 1 is a diagram showing a configuration of a navigation device to which the navigation device of the present invention is applied. A navigation device 100 shown in FIG. 1 is mounted on a vehicle, and includes a main body 10, a hard disk drive (HDD) 20, operation buttons 30, a communication unit 40, a GPS reception unit 50, an autonomous navigation sensor 60, a display 70, a speaker 80, A touch panel 90 and a microphone 92 are included.

  The HDD 20 is equipped with a hard disk 22 and reads map data recorded on the hard disk 22. The map data includes a plurality of leaf data corresponding to a map of a rectangular area at each stepwise scale. The figure data includes scale information and the longitude and latitude of the four corners of the rectangular area, and is uniquely identified by the scale, longitude, and latitude.

  The map data includes road information. The road information includes a set of a plurality of positions (nodes) on the road specified by latitude and longitude, and a kana character string (road name character string) of the road name. The map data includes regional information. This area information includes a set of longitude and latitude of the outer periphery of the area, the longitude and latitude of the representative point, and a kana character string of the place name.

  The map data recording medium is not limited to the hard disk 22, and may be a DVD, a CD-ROM, or the like, and a drive corresponding to the recording medium is prepared.

  The operation button 30 is operated by the user in setting a destination when searching for a guidance route. The communication unit 40 is, for example, a mobile phone, and communicates with the outside in order to acquire various information necessary for the navigation device 100. The GPS receiver 50 receives GPS signals from GPS satellites necessary for detecting the position of the vehicle. The autonomous navigation sensor 60 includes a gyro and the like for detecting the traveling direction of the vehicle, and a distance sensor that generates a pulse (vehicle speed pulse) at every constant travel distance.

  The display 70 displays various images such as a map image and an operation screen image. The touch panel 90 corresponds to the operation unit, is attached to the display surface of the display 70, and constitutes a touch panel type display unit together with the display 70. When the user's finger or the like touches, the touch panel 90 outputs information on the contact position. The microphone 92 detects the user's voice and generates and outputs a corresponding voice signal.

  The main body 10 includes a control unit 102, a buffer memory 104, a map drawing unit 106, an operation screen / mark drawing unit 108, a guidance route storage unit 110, a guidance route drawing unit 112, an image composition unit 118, a voice recognition dictionary memory 120, and voice recognition. The unit 122 is configured.

  The control unit 102 controls the entire navigation device 100. Specifically, the control unit 102 determines the position, direction, and position of the vehicle based on the GPS signal received by the GPS receiving unit 50, the traveling direction of the vehicle detected by the autonomous navigation sensor 60, and the output interval of the vehicle speed pulse. Calculate the speed. Further, the control unit 102 instructs the HDD 20 to read the map data recorded on the hard disk 22. The map data read out by the HDD 104 in response to an instruction from the control unit 102 is stored in the buffer memory 104. Further, the control unit 102 instructs the map drawing unit 106 to draw a map image. In response to this instruction, the map drawing unit 106 reads the map data stored in the buffer memory 104 and generates a map image based on the map data.

  Further, the control unit 102 issues an image drawing instruction to the operation screen / mark drawing unit 108 according to the operation of the operation button 30 by the user, the operation of the touch panel 90, the operation state of the navigation device 100, and the like. In response to this instruction, the operation screen / mark drawing unit 108 draws various operation screens, a vehicle position mark, and an image of the cursor position.

  Further, the control unit 102 reads out map data stored in the buffer memory 104 in accordance with an operation for searching the guidance route of the operation button 30 by the user, and executes a predetermined route search algorithm based on the map data. Use to set the guide route to the destination. Information on the set guide route (for example, a set of nodes (longitude and latitude) corresponding to the guide route from the departure point to the destination) is stored in the guide route storage unit 110. Further, the control unit 102 instructs the guidance route drawing unit 112 to draw an image of the guidance route at a predetermined timing such as the start of route guidance. In response to this instruction, the guidance route drawing unit 112 generates a guidance route image based on the guidance route information stored in the guidance route storage unit 110.

  Further, the control unit 102 instructs the image composition unit 118 to perform image composition in accordance with the operation of the operation button 30 by the user, the operation of the touch panel 90, the operation state of the navigation device 100, or the like. In response to this instruction, the image synthesis unit 118 appropriately synthesizes various images generated by the map drawing unit 106, the operation screen / mark drawing unit 108, and the guidance route drawing unit 112, and displays the obtained image on the display 70. Let

  In addition, the control unit 102 generates sound according to the operation of the operation button 30 by the user, the operation of the touch panel 90, the operation state of the navigation device 100, and the like, and outputs the sound to the speaker 80. For example, the control unit 102 generates guidance voice such as an instruction to turn at an intersection based on the guidance route information during route guidance, and causes the speaker 80 to output the guidance voice.

  In the navigation device 100 of the present embodiment, destination setting is performed by voice recognition. Prior to the destination setting by voice recognition, the control unit 102 generates place name recognition dictionary information and road name recognition dictionary information and stores them in the voice recognition dictionary information memory 120. Specifically, the control unit 102 instructs the HDD 20 to read out the area information and road information in the map data. In response to this instruction, the HDD 20 reads all area information and road information in the map data from the hard disk 22 and stores them in the buffer memory 104.

  The control unit 102 reads the area information and road information stored in the buffer memory 104. Next, the control part 102 extracts the place name character string in each area information, and produces | generates the place name speech recognition dictionary information which consists of these. Further, the control unit 102 extracts a road name character string in each road information, and generates road name speech recognition dictionary information including these. FIG. 2A shows the configuration of place name speech recognition dictionary information, and FIG. 2B shows the configuration of road name speech recognition dictionary information.

  Further, the control unit 102 stores the generated place name speech recognition dictionary information and road name speech recognition dictionary information in the speech recognition dictionary memory 120. The speech recognition dictionary memory 120 stores the place name speech recognition dictionary information and the road name speech recognition dictionary information.

  After the above-described place name recognition dictionary information and road name recognition dictionary information are generated and stored, destination setting is performed by voice recognition. FIG. 2 is a flowchart showing a destination setting operation by voice recognition.

  The control unit 102 determines whether or not the user has operated the operation button 30 to instruct voice recognition of the place name and road name of the destination (S101). When voice recognition of a destination place name and a road name is instructed, the control unit 102 displays an operation screen (character selection screen) for the user to select characters on the operation screen / mark drawing unit 108. Instruct to draw. In response to this instruction, the operation screen / mark drawing unit 108 draws an image of the character selection screen.

  The image of the character selection screen is output to the display 70 via the image composition unit 118 and displayed on the display. FIG. 4 is a diagram showing the configuration of the character selection screen. The character selection screen shown in FIG. 4 includes a button 201 for each character of Japanese syllabary, a display column 202 for a selected character or character string, and a confirmation button 203 for confirming the selected character or character string. Consists of images.

  Next, the control unit 102 determines whether or not a character or character string has been selected by the user (S103). Specifically, when the user touches the touch panel 90 with a character or the like being displayed on the display 70, the touch panel 90 outputs a contact position to the control unit 102. Based on the contact position and the positions of the character button 201 and the confirmation button 203 on the character selection screen, the control unit 102 selects which character button 201 the user has selected or selects the confirmation button 203. It can be determined whether or not.

  If one or more character buttons 201 are selected and then the confirm button 203 is selected, it is determined that a character or character string corresponding to the one or more character buttons 201 has been selected. When a character or a character string is selected, the control unit 102 generates the character or character string (S104). Here, the character string is an array of characters corresponding to the selection order of the character buttons 201.

  Next, the control unit 102 generates a place name character string including a generated character or a character string at the head of the place name character strings constituting the place name speech recognition dictionary information stored in the speech recognition dictionary memory 120, and the speech recognition dictionary. Of the road name character strings constituting the road name speech recognition dictionary information stored in the memory 120, a road name character string including the generated character or character string at the head is extracted (S105).

  Next, the control unit 102 instructs the voice recognition unit 122 to perform voice recognition. In response to this instruction, when the voice recognition unit 122 receives a voice signal corresponding to the user's voice sent from the microphone 92, the voice recognition unit 122 recognizes the user's voice and performs pattern information (a bit string corresponding to the user's voice). User voice pattern). Further, the voice recognition unit 122 generates a character string (user voice character string) corresponding to the user voice pattern (S106). The generated user voice character string is output to the control unit 102.

  Next, the control unit 102 searches the place name character string and road name character string extracted in S105 using the user voice character string as a key, and the user voice is extracted from the extracted place name character string and road name character string. The most similar character string is specified (S107). Here, for example, the control unit 102 identifies a place name character string and a road name character string extracted that completely match the user voice character string, and if there is no exact match, the user By specifying the process by gradually decreasing the degree of similarity, such as specifying a character that differs from the phonetic character string by one character, the user's phonetic character string is extracted from the extracted place name character string and road name character string. Identify the most similar.

  Next, the control unit 102 generates a sound pattern corresponding to the place name character string or the road name character string specified in S107, and further generates a sound signal corresponding to the sound pattern. Further, the control unit 102 causes the speaker 80 to output sound corresponding to the generated sound signal (S108).

  Thereafter, the control unit 102 determines whether or not the place name or road name corresponding to the output voice has been determined as the destination by the user (S109). For example, when the user presses a confirmation button (not shown) of the operation button 30, the control unit 102 determines that the place name or road name corresponding to the output voice has been decided as the destination.

  When the place name or road name is confirmed as the destination, the control unit 102 sets the decided place name or road name as the destination (S110). Specifically, the control unit 102 includes, in the area information and road information stored in the buffer memory 104, the area information including the place name character string corresponding to the confirmed place name or the road name character corresponding to the confirmed road name. Identify road information including columns. Furthermore, when specifying the area information, the control unit 102 sets the longitude and latitude of the representative point in the area information as the destination, and when specifying the road information, the control unit 102 sets a plurality of points in the road information. The latitude and longitude of any one of the nodes (for example, the longitude and latitude of the node closest to the vehicle position) is set as the destination. Thereafter, a guidance route to the set destination is searched.

  On the other hand, when the place name or road name corresponding to the output voice is not determined by the user as the destination, for example, when the user presses a cancel button (not shown) of the operation button 30, control is performed. The unit 102 excludes the place name character string or the road name character string specified in S107 from the target when the place name character string or the road name character string is specified in S107 (S111). Thereafter, the operations after S107 are repeated, and the control unit 102 identifies the most similar place name character string and road name character string to the user voice character string (S107).

  In the embodiment described above, the control unit 102 corresponds to an acquisition unit, a character string extraction unit, and a character string search unit, and the speech recognition dictionary memory 120 corresponds to a storage unit. The voice recognition unit 122 corresponds to a voice recognition unit, the speaker 80 corresponds to an output unit, and the operation screen / mark drawing unit 108 corresponds to a display processing unit.

  As described above, in the navigation device 100 of the present embodiment, when the user selects a character or a character string by touching the touch panel 90 with a finger or the like while the character selection screen is displayed on the display 70, the control unit 102 Is a character or character string selected by the user among the place name character string constituting the place name voice recognition dictionary information and the road name character string constituting the road name voice recognition dictionary information stored in the voice recognition dictionary memory 120. The place name character string and road name character string included in the head are extracted. Thereafter, when the user utters a voice, the voice recognition unit 122 generates a user voice character string corresponding to the voice, and the control unit 102 selects the user among the extracted place name character string and road name character string. Identify the most similar to the phonetic string. A place name corresponding to the specified place name character string or a road name corresponding to the road name character string is a destination candidate.

  Therefore, according to the selection of the user's character or character string, the place name character string constituting the place name speech recognition dictionary information and the road name character string constituting the road name speech recognition dictionary information are further narrowed down, and then the voice Location name speech recognition dictionary information and road name speech recognition dictionary information by recognition and destination candidate search by speech recognition are simultaneously performed, and place name character strings constituting location name speech recognition dictionary information to be searched And the amount of road name character strings constituting the road name speech recognition dictionary information is reduced, and the certainty of search is improved. Further, the narrowing of the place name character string constituting the place name speech recognition dictionary information and the road name character string constituting the road name speech recognition dictionary information is performed by selecting the character or character string selected by the user by touching the touch panel 90 with a finger or the like. This is based on this, and since speech recognition is not used when narrowing down as in the prior art, errors are unlikely to occur and the certainty of search is improved.

  In the above-described embodiment, the sound of the place name or road name corresponding to the place name character string or the road name character string specified in S107 is output in S108 of FIG. 3, but may be displayed on the display 70. Good. Further, in the above-described embodiment, the place name character string and the road name character string extracted from the place name speech recognition dictionary information and the road name speech recognition dictionary information are identified as being most similar to the user speech character string. A plurality of place name character strings and road name character strings similar to the person voice character string may be specified. In this case, for example, a place name character string and a road name character string that completely match the user voice character string, and a place name character string and a road name character string that are different by one character are specified.

  Further, in the above-described embodiment, the setting by the voice recognition of the destination when the destination is the setting point has been described. However, other setting points, for example, a waypoint through which the destination is reached The present invention can be similarly applied to settings by voice recognition.

  As described above, the navigation device according to the present invention can improve the certainty of searching for place names and road names based on voice recognition, and is useful as a navigation device.

It is a figure which shows the structure of a navigation apparatus. It is a figure which shows the structure of place name speech recognition dictionary information and road name speech recognition dictionary information. It is a flowchart which shows the operation | movement of the destination setting by speech recognition. It is a figure which shows an example of a character selection screen.

Explanation of symbols

10 Main body 20 Hard disk drive (HDD)
DESCRIPTION OF SYMBOLS 22 Hard disk 30 Operation button 40 Communication part 50 GPS receiving part 60 Autonomous navigation sensor 70 Display 80 Speaker 90 Touch panel 92 Microphone 100 Navigation apparatus 102 Control part 104 Buffer memory 106 Map drawing part 108 Operation screen and mark drawing part 110 Guidance path memory | storage part 112 Guide route drawing unit 118 Image composition unit 120 Speech recognition dictionary memory 122 Speech recognition unit

Claims (5)

Storage means for storing character strings of place names and road names that can be set points;
An acquisition means for acquiring a character or a character string selected by an operation of an operation unit by a user;
A character string extracting unit that extracts a character string or a character string acquired by the acquiring unit from a character string of a place name and a road name stored in the storage unit;
Voice recognition means for recognizing the voice emitted by the user and generating a recognition result;
Navigation having character string search means for searching a character string of a place name or road name corresponding to a recognition result generated by the voice recognition means among character strings of place names and road names extracted by the character string extraction means. apparatus. The voice recognition means generates a character string corresponding to the voice uttered by the user as a recognition result,
The character string search means includes a place name that is most similar to a character string corresponding to a voice generated by the user and generated by the voice recognition means among character strings of place names and road names extracted by the character string extraction means. The navigation device according to claim 1, wherein a character string of a road name is specified.   The navigation apparatus according to claim 1, further comprising an output unit that outputs a character string of a place name or a road name searched by the character string search unit.   The navigation device according to claim 3, wherein the output unit outputs a voice corresponding to a character string of a place name or a road name searched by the character string search unit. Display processing means for displaying a screen for character selection on a touch panel type display unit;
The navigation device according to claim 1, wherein the acquisition unit acquires a character or a character string selected by an operation of touching the display unit by the user.

Images