JP2013054074A - Speech recognition device, speech recognition method, and speech recognition program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a speech recognition device, a speech recognition method, and a speech recognition program that are capable of improving operability of a speech recognition command.SOLUTION: A navigation device 1 comprises: a speech detection unit 2b for detecting a user's speech utterance; a speech recognition command determination unit 2c for determining a speech recognition command contained in the user's speech utterance; and a speech recognition control unit 2a for executing a predetermined control corresponding to a speech recognition command determined by the speech recognition command determination unit 2c and for selecting a speech recognition command for display on a display 6. The speech recognition control unit 2a displays, on the display 6, a speech recognition command selected on the basis of a predetermined condition to determine whether or not the command is likely to be used by a user.

Description

  The present invention relates to a voice recognition device, a voice recognition method, and a voice recognition program.

  2. Description of the Related Art Conventionally, in a car navigation system or the like, a voice recognition device that can accept input of various voice recognition commands by recognizing a user's voice has been used.

  Such voice recognition commands are generally hierarchized by a hierarchical structure (tree structure). First, only a plurality of voice recognition commands positioned in the highest hierarchy are displayed on the display, and when any one of the plurality of voice recognition commands is received, the voice recognition command that has received the input is received. On the other hand, only a plurality of voice recognition commands positioned in the lower hierarchy are displayed on the display. Thereafter, the display and reception of the voice recognition command are repeated in the same manner, so that the hierarchy is lowered one step at a time, and finally one voice recognition command positioned at the lowest hierarchy is specified. By performing such processing, the number of voice recognition commands displayed on the display at a time can be limited, and the voice recognition rate by the voice recognition apparatus can be improved.

  However, when the total number of voice recognition commands increases as the number of functions of car navigation systems and the like increases, the hierarchy of voice recognition commands also deepens, and the user must input before finally specifying one voice recognition command. Since the number of voice recognition commands that have to be increased increases, there is a problem in that the operation is troublesome. In particular, in the case of a device installed in a vehicle such as a car navigation system, operation items may be restricted during the traveling of the vehicle from the viewpoint of safety. For this reason, when driving is started at the time when operation is performed halfway through multiple hierarchies, it is limited to perform further operations, and eventually one voice recognition command cannot be specified. May be wasted.

  In view of these points, proposals have also been made to improve the operability of input devices using voice recognition. For example, in the moving body input device, when the moving body is in a running state, it is determined whether or not the input received by the input unit is an item that can be executed by manual operation and voice, and the item operation by manual operation and voice is performed. It has been proposed to display on a display unit a display mode in which execution is possible (see, for example, Patent Document 1).

International Publication No. 2007/0669573 Pamphlet

  However, although the conventional apparatus as described above can grasp the operation items that can be executed in the running state, it essentially solves the problem caused by the voice recognition commands being hierarchized in a deep hierarchy. I couldn't. That is, in the above-described conventional apparatus, the required function can still be executed unless one voice recognition command positioned at the lowest hierarchy is specified by moving down the voice recognition command hierarchy step by step. It was not possible, and it took time and effort to operate.

  The present invention has been made in view of the above, and an object thereof is to provide a speech recognition device, a speech recognition method, and a speech recognition program that can improve the operability of speech recognition commands.

  In order to solve the above-described problems and achieve the object, the speech recognition apparatus according to claim 1 includes a voice detection unit that detects a user's speech, and a user's speech that is detected by the voice detection unit. Based on the voice recognition command specifying means for specifying the voice recognition command included in the speech voice of the user, and executing predetermined control corresponding to the voice recognition command specified by the voice recognition command specifying means, and Voice recognition control means for selecting and displaying a voice recognition command on the display means, the voice recognition control means selected based on a predetermined condition for determining whether or not the user is likely to use The voice recognition command is displayed on the display means.

  The voice recognition device according to claim 2 is the voice recognition device according to claim 1, wherein the voice recognition control means may be used by the user when a map is displayed on the display means. A voice recognition command selected based on a predetermined condition for determining whether or not is high is superimposed on the map and displayed on the display means.

  According to a third aspect of the present invention, there is provided the voice recognition apparatus according to the first or second aspect, further comprising voice output means for outputting voice to the user, wherein the voice recognition control means receives the voice recognition command. When displayed on the display means, the voice output means is made to output a voice output for allowing the user to recognize the displayed voice recognition command.

  The voice recognition device according to claim 4 is the voice recognition device according to claim 3, wherein the voice recognition control means displays the voice recognition commands on the display means when the plurality of voice recognition commands are displayed. Priorities among the voice recognition commands are determined based on a predetermined priority determination condition, and voice output for causing the user to recognize the voice recognition commands in descending order of the determined priorities is the voice output. Let the means output.

  Further, the voice recognition device according to claim 5 is a voice recognition device according to any one of claims 1 to 4, wherein usage history information relating to a history of usage status of a voice recognition command by a user is stored. A situation history information storage means, wherein the voice recognition control means displays the voice recognition command on the display means and then uses the voice recognition command usage situation history information based on the usage situation of the voice recognition command by the user. Is stored in the usage status history information storage means, and the voice recognition command is selected based on the usage status history information stored in the usage status history information storage means when the voice recognition command is displayed on the display means.

  According to a sixth aspect of the present invention, there is provided a voice recognition method for detecting a user's utterance voice, and a voice included in the user's utterance voice based on the user's utterance voice detected in the voice detection step. A voice recognition command specifying step for specifying a recognition command; and a voice for executing predetermined control corresponding to the voice recognition command specified in the voice recognition command specifying step and selecting and displaying the voice recognition command on the display means A voice recognition control step, wherein the voice recognition command selected based on a predetermined condition for judging whether or not the user is likely to use is displayed on the display means. .

  A voice recognition program according to a seventh aspect causes a computer to execute the voice recognition method according to the sixth aspect.

  According to the speech recognition device according to claim 1, the speech recognition method according to claim 6, and the speech recognition program according to claim 7, a predetermined determination as to whether or not the user is likely to use is performed. Since the voice recognition command selected based on the condition is displayed on the display means, for example, when a plurality of voice recognition commands are hierarchized in a hierarchical structure, the user can display the plurality of voice recognition commands hierarchically. Since voice recognition commands that are likely to be used now can be used directly without descending sequentially from the upper hierarchy to the lowest hierarchy, the operability of hierarchical voice recognition commands can be improved.

  According to the voice recognition device of the second aspect, when the map is displayed on the display unit, the voice recognition selected based on a predetermined condition for judging whether or not the user is likely to use the map. Since the command is displayed superimposed on the map, it is possible to visually notify the user of voice recognition commands that are currently available even during map display, and the user viewing the map uses the voice recognition command. Convenience can be improved.

  Further, according to the voice recognition device of the third aspect, since the voice recognition command displayed on the display means is output as voice, the voice recognition command that can be currently used can be notified to the user by voice, It is possible to improve convenience and safety when the user uses the voice recognition command.

  According to the voice recognition device of the fourth aspect, the priority among the plurality of voice recognition commands is determined, and the voice recognition commands are output in the order of the determined priority. The voice recognition commands that can be used can be grasped from the highest priority order. For example, the voice recognition commands with the higher priority order can be used without waiting for the voice recognition commands with the lower priority order to output the voice. It is possible to further improve convenience when the user inside uses the voice recognition command.

  According to the voice recognition device of the fifth aspect, since the voice recognition command is selected based on the usage history information, it is possible to select a voice recognition command that matches the user's past usage status, The convenience of the user can be further improved by increasing the selection accuracy of the recognition command.

It is a block diagram which illustrates a navigation device concerning an embodiment of the invention. It is the table | surface which illustrated the speech recognition command table. It is the table | surface which illustrated the usage status history information stored in usage status history information DB. It is a flowchart of a voice recognition control process. It is a flowchart of an adaptive matching process. It is a flowchart of a map superimposition display process. It is a display example of a map in which a superimposed display command is superimposed. It is a flowchart of a speech recognition command execution process. It is a flowchart of the same screen display process. 10 is a flowchart of the same screen display process continued from FIG. 9. It is a display example of the same screen display command as the standard command. It is the figure which illustrated the hierarchical structure of the voice recognition command.

  Hereinafter, embodiments of a speech recognition device, a speech recognition method, and a speech recognition program according to the present invention will be described in detail with reference to the drawings. However, the present invention is not limited by this embodiment.

(Basic concept)
The speech recognition device can be configured as a single device or as part of another device or system. Below, the example which comprised the speech recognition apparatus as a part of navigation apparatus mounted in the vehicle is demonstrated.

The “voice recognition command” is an operation instruction for various connected devices such as a navigation apparatus and an air conditioner connected to the voice recognition apparatus, and is an operation instruction that can be specified by the uttered voice.
This voice recognition command is hierarchized by a hierarchical structure (tree structure), for example, as in the prior art (associated with the hierarchical structure). Of each hierarchy in this hierarchical structure, the highest hierarchy is the "highest hierarchy", the lowest hierarchy is the "lowest hierarchy", and the hierarchy between these highest and lowest hierarchy is the "middle hierarchy" Called. In addition, the voice recognition command positioned at the highest hierarchy is “highest hierarchy command”, the voice recognition command positioned at the lowest hierarchy is “lowest hierarchy command”, and the voice recognition command positioned at the middle hierarchy is “ This is referred to as a “middle hierarchy command”.

  A part of the plurality of voice recognition commands hierarchized in this way is selected and displayed on the display. In this embodiment, as a selection method when selecting a voice recognition command to be displayed on the display in this way, in the present embodiment, a selection method based on a hierarchical structure and a user (mainly a driver of a vehicle). , A passenger may be used), and a method of selecting based on a predetermined condition for determining whether or not there is a high possibility of use is used. Hereinafter, the selection by the former method is referred to as “standard selection”, and the selection by the latter method is referred to as “adaptive selection”. The voice recognition command selected by the standard selection is referred to as “standard command (first voice recognition command)”, and the voice recognition command selected by adaptive selection is referred to as “adaptive command (second voice recognition command)”.

  As a method for displaying an adaptive command on the display, in the present embodiment, when a map is displayed on the display, a method for displaying the adaptive command superimposed on the map and a case where a standard command should be displayed on the display. Two methods are used: an adaptive command is displayed simultaneously with the standard command. Displaying simultaneously means displaying the standard command and the adaptive command on the same screen in the display. Hereinafter, the adaptive command displayed by the former method is referred to as “superimposed display command”, and the adaptive command displayed by the latter method is referred to as “same screen display command”.

(Configuration-navigation device)
First, a configuration of a navigation device that includes the voice recognition device according to the present embodiment as a part thereof will be described. FIG. 1 is a block diagram illustrating a navigation device 1 according to this embodiment. As shown in FIG. 1, the navigation device 1 includes a control unit 2 and a data recording unit 3. The navigation device 1 is connected to a microphone 4, a touch panel 5, a display 6, a speaker 7, and a current position detection processing unit 8.

(Configuration-Navigation device-Control unit)
The control unit 2 is a control unit that controls the navigation device 1. Specifically, the CPU, various programs that are interpreted and executed on the CPU (a basic control program such as an OS, a specific function that is activated on the OS, and the like) And an internal memory such as a RAM for storing the program and various data. In particular, the voice recognition program according to the present embodiment is substantially installed in the navigation device 1 via an arbitrary recording medium or network, thereby substantially configuring each unit of the control unit 2.

  The control unit 2 includes a voice recognition control unit 2a, a voice detection unit 2b, and a voice recognition command specifying unit 2c in terms of functional concept. The voice recognition control unit 2a is a voice recognition control unit that executes predetermined control corresponding to the voice recognition command specified by the voice recognition command specifying unit 2c, and selects the voice recognition command to be displayed on the display 6. The voice detection unit 2b is a voice detection unit that detects the voice of the user. The voice recognition command specifying unit 2c is a voice recognition command specifying unit that specifies a voice recognition command included in the user's uttered voice based on the user's uttered voice detected by the voice detecting unit 2b. Details of processing executed by each of these units will be described later.

(Configuration-Navigation device-Data recording unit)
The data recording unit 3 is a recording unit that records a program and various data necessary for the operation of the navigation device 1, and is configured using a magnetic recording medium such as a hard disk (not shown) as an external storage device, for example. ing. However, any other recording medium including a semiconductor storage medium such as a flash memory or an optical recording medium such as a DVD or a Blu-ray disk can be used instead of or together with the hard disk.

  The data recording unit 3 includes a map information database (hereinafter referred to as “DB”) 3a, a voice recognition command table 3b, and a usage status history information DB 3c.

  The map information DB 3a is map information storage means for storing map information. “Map information” includes, for example, link data (link number, connection node number, road coordinates, road type, number of lanes, legal speed, travel regulations, etc.), node data (node numbers, coordinates), feature data (signals, roads Signs, guardrails, buildings, etc.), facility data (position of each facility, type of each facility, etc.), terrain data, map display data for displaying a map on the display 6, and the like.

  The voice recognition command table 3b is a table that stores information (information for adaptive selection) necessary for selecting a voice recognition command by adaptive selection. As illustrated in FIG. 2, the voice recognition command table 3b includes an item “voice recognition command”, an item “function classification”, an item “selection criterion classification”, an item “user ID”, an item “priority”, and an item. “Display algorithm” is included, and information corresponding to these items is associated with each other. Information corresponding to the item “voice recognition command” is command information for specifying the voice recognition command. However, here, only the lowest hierarchy command (eg, “regulation information” in FIG. 2) is included in the voice recognition command. The information corresponding to the item “function classification” is classification information when the voice recognition command is classified from the functional aspect (for example, “information display” in FIG. 2). The information corresponding to the item “selection reference classification” is classification information when the voice recognition command is classified from the selection reference plane in the case of adaptive selection. There are three types of information: “situation adaptive type”, “environment adaptive type”, and “user adaptive type”. The “situation adaptive type” is a classification to which a voice recognition command selected according to a short-term situation (for example, traffic situation) belongs. “Environment-adaptive type” is a classification to which a voice recognition command selected according to a medium- to long-term situation (for example, traveling on a highway) belongs. The “user adaptive type” is a classification to which a voice recognition command selected based on a user attribute, behavior tendency, behavior history, or the like belongs. The information corresponding to the item “user ID” is user identification information for uniquely identifying the user of the vehicle (for example, “UID0001” in FIG. 3). The information corresponding to the item “priority” is a priority determination condition for determining the priority among the plurality of adaptation commands when the plurality of adaptation commands are displayed on the display 6. In the example of FIG. 2, the priority is indicated by numerical values from “1” to “5”, and the lower the numerical value, the higher the priority. In particular, the priority of an adaptive command in which selection criterion classification = user adaptive type is stored for each user ID. Information corresponding to the item “display algorithm” is a predetermined condition (selection condition) for determining whether or not the user is highly likely to use the information. For example, the example of the top record in FIG. 2 shows that the item “speech recognition command” = when the condition “when traveling on an expressway and traffic regulation information on the travel route is received” is met. This indicates that “regulation information” is selected as an adaptation command. Such a voice recognition command table 3b is determined based on an arbitrary standard and is recorded in the data recording unit 3 when the navigation device 1 is manufactured. However, the priority corresponding to the selection criterion classification = user-adaptive voice recognition command is updated as necessary in a voice recognition command execution process described later.

  The usage status history information DB 3c is usage status history information storage means for storing information related to the usage status history by the user of the voice recognition command (hereinafter referred to as usage status history information). As illustrated in FIG. 3, the usage history information includes an item “user ID”, an item “voice recognition command”, an item “departure point”, an item “destination”, an item “current position”, and an item “guidance start”. "Date and time", item "current date and time", item "travel road type", item "vehicle speed", and item "gasoline remaining amount", and information corresponding to each of these items is configured to be associated with each other . The information corresponding to the item “user ID” is user identification information for uniquely identifying the user of the vehicle (for example, “UID0001” in FIG. 3). The information corresponding to the item “voice recognition command” is the same as the information of the same name in the voice recognition command table 3b of FIG. The information corresponding to the item “departure place” is information for specifying the departure place of the vehicle set in the navigation device 1 (in FIG. 3, the coordinates are, for example, “35 degrees 35 minutes 21 seconds, 139 Degree 38 minutes 18 seconds "). The information corresponding to the item “Destination” is information for specifying the destination of the vehicle set in the navigation device 1 (in FIG. 3, it is a coordinate, for example, “35 degrees 40 minutes 12 seconds, 139 Degree 12 minutes 10 seconds "). The information corresponding to the item “current position” is information for specifying the position of the vehicle when the usage history information is acquired (in FIG. 3, the coordinates are, for example, “35 degrees 37 minutes 12 seconds, 139 degrees 20 minutes 01 seconds "). The information corresponding to the item “guidance start date / time” is information for specifying the date / time when the navigation device 1 started the guidance of the travel route (in FIG. 3, the date (year display) and the time (24 hours display) are displayed. For example, “2011/08/01 13:55”). The information corresponding to the item “current date and time” is information for specifying the date and time when the usage history information is acquired (in FIG. 3, the date (year display) and the time (24 hours display) are written together. For example, “2011/08/01 14:43”). The information corresponding to the item “traveling road type” is information for specifying the type of road on which the vehicle is traveling when the usage history information is acquired (in FIG. 3, for example, “highway”) ). The information corresponding to the item “vehicle speed” is information for specifying the vehicle speed of the vehicle when the usage history information is acquired (for example, “100 Km / h” in FIG. 3). The information corresponding to the item “remaining gasoline” is information for specifying the remaining amount of gasoline in the vehicle when the usage history information is acquired (for example, “34L” in FIG. 3). Such usage status history information in the usage status history information DB 3 c is updated and recorded in the data recording unit 3 in a voice recognition command execution process described later.

(Configuration-microphone)
Returning to FIG. 1, the microphone 4 converts the sound into an electrical signal and outputs it to the navigation device 1. As this microphone 4, a known microphone can be used.

(Configuration-touch panel)
The touch panel 5 receives various manual inputs from the user when pressed by the user's finger or the like. The touch panel 5 is formed to be transparent or translucent, and is provided on the front surface of the display 6 so as to overlap the display surface of the display 6. As the touch panel 5, for example, a publicly known touch panel provided with operation position detecting means by a resistance film method or a capacitance method can be used.

(Configuration-Display)
The display 6 is display means for outputting information based on the control of the navigation device 1. In particular, the display 6 is a display unit that displays a voice recognition command selected by the voice recognition control unit 2a from a plurality of predetermined voice recognition commands hierarchized by a hierarchical structure. The specific configuration of the display 6 is arbitrary, and a known flat panel display such as a liquid crystal display or an organic EL display can be used.

(Configuration-Speaker)
The speaker 7 is a voice output unit that outputs information by voice based on the control of the navigation device 1. In particular, when the voice recognition command is displayed on the display 6, the speaker 7 is voice output means for outputting a voice output for allowing the user to recognize the displayed voice recognition command. The specific form of the sound output from the speaker 7 is arbitrary, and it is possible to output a synthesized sound generated as necessary and a sound recorded in advance.

(Configuration-Current position detection processing unit)
The current position detection processing unit 8 is current position detection means for detecting the current position of the vehicle on which the navigation device 1 is mounted. Specifically, the current position detection processing unit 8 includes at least one of a GPS, a geomagnetic sensor, a distance sensor, and a gyro sensor (all of which are not shown), and determines the current vehicle position (coordinates), direction, and the like. Detection is performed by a known method.

(Processing-Voice recognition control process)
Next, a speech recognition control process executed by the navigation device 1 configured as described above will be described. FIG. 4 is a flowchart of the voice recognition control process (in the following description of each process, step is abbreviated as “S”). This voice recognition control process is started after the navigation apparatus 1 is powered on, for example.

  As shown in FIG. 4, when the voice recognition control process is started, the voice recognition control unit 2a acquires a user ID (SA1). For example, when a message prompting the user to input his / her user ID is displayed on the display 6 or voice output from the speaker 7 and the user inputs his / her user ID via the touch panel 5, the input Obtained user ID. In addition, the user ID can be acquired by various known methods. For example, the user's face is photographed with a camera attached to the inside of the vehicle, and the user is recognized based on the photographed image. Thus, the user is identified, and the user ID corresponding to the identified user is acquired from the user ID recorded in advance in the data recording unit 3 in association with the user image.

(Processing-Voice recognition control processing-Adaptive matching processing)
Next, the speech recognition control unit 2a activates adaptive matching processing (SA2). FIG. 5 is a flowchart of the adaptive matching process. In this process, the voice recognition control unit 2a acquires various information (SB1). As the various types of information, information necessary for determining the sufficiency of the display algorithm of the voice recognition command table 3b in FIG. 2 is acquired. For example, the voice recognition control unit 2a acquires a departure place, a destination, a current position, a guidance start date and time, a current date and time, a traveling road type, a vehicle speed, and a gasoline remaining amount. Among these, the departure point, the destination, the current position, the guidance start date and time, the current date and time, and the travel road type are acquired from information set in the navigation device 1 by the user or from the travel route set by the navigation device 1. It can be acquired or acquired from the map information DB 3a. The vehicle speed can be acquired from a vehicle speed sensor (not shown) mounted on the vehicle, and the gasoline remaining amount can be acquired from a combustion tank sensor (not shown) mounted on the vehicle. Further, the voice recognition control unit 2a receives traffic regulation information and traffic jam information on the travel route from a traffic center, a beacon, or the like via a communication unit (not shown). In addition, it is possible to acquire not only current temporary information but also information with a certain time lapse. For example, the average speed for X minutes continues to acquire the acquired vehicle speed at predetermined intervals for X minutes, It can be obtained by calculating the average value.

  Further, the voice recognition control unit 2a acquires usage status history information (SB2). Specifically, the voice recognition control unit 2a acquires usage status history information corresponding to the user ID acquired in SA1 of FIG. 4 from the usage status history information stored in the usage status history information DB 3c.

  Then, the voice recognition control unit 2a selects an adaptation command (SB3). Specifically, the voice recognition control unit 2a is satisfied with various information acquired in SB1 from the voice recognition commands of selection criteria classification = environment adaptive type or situation adaptive type in the voice recognition command table 3b of FIG. The voice recognition command corresponding to the display algorithm is specified, and the specified voice recognition command is selected as the adaptive command. For example, when the travel road type = highway is acquired as the various information acquired in SB1, and traffic regulation information on the travel route is received, the voice recognition command = display algorithm “high speed” Since “when traffic regulation information on a travel route is received while traveling on a road” is satisfied, voice recognition command = regulation information is set as a candidate for an adaptation command. In addition, the voice recognition control unit 2a specifies a voice recognition command corresponding to a display algorithm that matches the usage history information acquired in SB2 from the voice recognition commands of selection criterion classification = user adaptive type, and The selected voice recognition command is selected as an adaptive command. For example, based on the usage history information acquired in SB2, when the current user is a user who often sets general road guidance immediately after the start of guidance (for example, guidance start as a record corresponding to the user ID of the user) When the elapsed time from the date and time to the current date and time is within a predetermined time and the number of records in which the voice recognition command is general road priority guidance is included in a predetermined number or more), voice recognition command = general road priority guidance , To be candidates for adaptive commands.

  In SB3, the speech recognition control unit 2a selects a predetermined number (for example, three) of adaptive commands. For example, when the number of selected adaptation commands is less than a predetermined number, the adaptation with the highest priority stored in the speech recognition command table from among the speech commands not yet selected from the speech recognition command table 3b. A command may be selected. On the other hand, if the selected number of adaptive commands exceeds the predetermined number, the predetermined number of adaptive commands may be selected from the selected adaptive commands in descending order of priority stored in the voice recognition command table. Alternatively, a predetermined number of adaptive commands may be selected at random. Further, the same number of adaptive commands may be selected for each function classification for the predetermined number of adaptation commands (for example, when the predetermined number = 4, function recognition = information recognition voice recognition command, function classification = A voice recognition command for switching map display, a voice recognition command for function classification = operation, and a voice recognition command for function classification = guidance are selected one by one). Alternatively, for the predetermined number of adaptation commands, the same number of adaptation commands may be selected for each selection criterion classification (for example, when the predetermined number = 3, the selection criterion classification = speech recognition command of environment adaptation type, Selection criteria classification = situation-adaptive voice recognition command, selection criteria classification = personal adaptation-type voice recognition command, etc., one by one).

  Next, the speech recognition control unit 2a determines the priority order between the adaptive commands selected in SB3 based on a predetermined priority order determination condition (SB4). The priority order is a priority order for displaying the adaptive command on the display 6 and also a priority order for outputting the adaptive command by voice. Moreover, the priority of the voice recognition command table 3b is applied as the priority order determination condition. Specifically, the voice recognition control unit 2a acquires the priority of the adaptive command selected in SB3 from the voice recognition command table 3b, and sorts the adaptive commands in descending order of the acquired priority. For example, in SB3, when restriction information (priority = 1), city map display (priority = 3), and traffic jam information (priority = 2) are selected as adaptation commands, these are sorted and restricted. Sort in the order of information, traffic jam information, and city map display. At this time, the priority corresponding to the user ID acquired in SA1 in FIG. 4 is used as the priority of the adaptation command in which selection criterion classification = user adaptive type.

  Next, the voice recognition control unit 2a determines whether there is an update of the adaptive commands sorted in SB4 (SB5). That is, the data recording unit 3 displays the adaptive command currently displayed on the display 6 (or, if not displayed at that time, the last displayed command) and the sorting of the adaptive commands. Results are recorded. Then, the voice recognition control unit 2a determines whether or not the adaptive commands sorted in SB4 and the sorting results are partially changed from the adaptive commands recorded in the data recording unit 3 and the sorting results. If it has not been changed, it is assumed that the adaptive command has not been updated (SB5, No), and the adaptive matching process is terminated. On the other hand, if it has been changed, it is assumed that there is an update of the adaptation command (SB5, Yes), and an update flag indicating that the adaptation command has been updated is turned ON and recorded in the data recording unit 3 (SB6). The matching process ends.

(Processing-Voice recognition control process)
Next, returning to the voice recognition control process of FIG. 4, the voice recognition control unit 2a determines whether or not a map is being displayed on the display 6 (SA3). When the display is being performed (SA3, Yes), the voice recognition control unit 2a activates a map superposition display process for superimposing and displaying a superimposition display command on the map (SA4). On the other hand, when not being displayed (SA3, No), the voice recognition control unit 2a determines whether a menu for selecting a voice identification command (hereinafter referred to as a voice identification menu) is being displayed on the display 6 (SA5). ) When the display is in progress (SA5, Yes), the same screen display process for displaying the same screen display command is started (SA6). When the display is not in progress (SA5, No), these map superpositions are performed. The process proceeds to SA7 without starting the display process or the same screen display process.

(Processing-Voice recognition control process-Map overlay display process)
First, the map overlay display process (SA4) will be described. FIG. 6 is a flowchart of the map overlay display process. In this process, the voice recognition control unit 2a displays a superimposed display command on the map displayed on the display 6 (SC1). That is, the voice recognition control unit 2a displays the adaptive commands selected in SB3 of the adaptive matching process in FIG. 5 and sorted in SB4 as superimposed display commands. Since this superimposed display command is a voice recognition command (lowest hierarchy command) positioned in the lowest hierarchy in the hierarchical structure, a voice recognition command that allows the user to directly specify the operation content can be displayed. It becomes possible to quickly perform an operation content instruction by voice recognition by performing a shortcut.

  Thus, it is preferable that the display area when the superimposed display command is displayed in a superimposed manner is a display area excluding the display area considered to be important for the user. The display area considered to be important for the user includes, for example, an area in which the current position of the vehicle is displayed, and an area in which the departure point, destination, and travel route are displayed when a travel route is set. An enlarged display area of an intersection or a guide sign, or a schematic display area such as a high-speed schematic diagram. FIG. 7 shows a display example of a map on which the superimposed display command is superimposed. This display example shows a case where the high-speed schematic diagram 11 is superimposed on the map 10, and a plurality of superimposed display commands are displayed in the display area of the map 10 excluding the display area of the high-speed schematic diagram 11. The superimposed display command menu 13 constituted by 12 is displayed in a superimposed manner. When the superimposed display commands 12 are displayed in the vertical arrangement in this way, it is preferable to display the commands in order of decreasing priority from top to bottom. Alternatively, the superimposed display commands 12 may be superimposed and displayed side by side, and in this case, it is preferable to display the commands in order of decreasing priority from left to right. In either case of vertical alignment or horizontal alignment, the superimposed display command 12 is preferably displayed along the outer frame in the vicinity of the outer frame of the display area.

  Next, in FIG. 6, the voice recognition control unit 2a determines whether or not the update flag of the data recording unit 3 is ON (SC2). If the updated flag is ON (SC2, Yes), the voice recognition control unit 2a turns OFF the updated flag in the data recording unit 3 (SC3) and superimposes superimposed on the map at SC1. The display command is output as a voice in order of priority (SC4). Specifically, the superimposed display commands are output through the speaker 7 in descending order of priority (order sorted by SB4 in the adaptive matching process in FIG. 5). For example, in the case of FIG. 7, a voice message such as “congestion information, high-speed schematic diagram, restriction information can be used as a voice recognition command” is output. Thus, the user can grasp the currently available adaptation commands in order of priority, without looking at the display 6.

  Next, in FIG. 6, after performing voice output (SC4) or when the update flag is not ON (SC2, No), the voice recognition command specifying unit 2c is within a predetermined time (for example, 1 minute). Then, it is determined whether or not an operation of a superimposed display command has been performed (SC5). Specifically, the voice recognition command specifying unit 2c monitors whether or not the user's uttered voice is input through the microphone 4 and is detected by the voice detection unit 2b. Whether the uttered voice includes any one of the superimposed display commands currently displayed on the display 6 is recognized by a known voice recognition method. If the superimposed display command is included, the voice recognition command specifying unit 2c determines that the voice recognition command has been operated (SC5, Yes). In this case, since the superimposed display command is a voice recognition command positioned at the lowest hierarchy, and the operation content can be immediately specified when the superimposed display command is operated, the voice recognition command targeted for the superimposed display command. The execution process is started (SC6). When the voice recognition command execution process ends or when it is determined that there is no operation for the superimposed display command (No in SC5), the map superimposed display process ends.

(Processing-Voice recognition control process-Map overlay display process-Voice recognition command execution process)
Next, the voice recognition command execution process (SC6) will be described. FIG. 8 is a flowchart of voice recognition command execution processing. In this process, the voice recognition control unit 2a executes a predetermined operation content corresponding to the superimposed display command determined to be operated in SC5 of FIG. 6 (SD1). For example, when the superimposed display command = regulation information, the regulation information acquired through the communication unit is displayed on the display 6.

  Next, the voice recognition control unit 2a acquires various information (SD2). As the various types of information, each information excluding the user ID and the voice recognition command is acquired from the usage status history information in the usage status history information DB 3c of FIG. These pieces of information can be obtained in the same manner as SB1 in FIG. Next, the voice recognition control unit 2a updates the usage status history information (SD3). Specifically, the voice recognition command that is the superimposed display command executed in SD1, the various information acquired in SD2, and the user ID acquired in SA1 in FIG. The new record is stored in the usage status history information DB 3c.

  Thereafter, the voice recognition control unit 2a updates the priority as necessary (SD4). Specifically, the voice recognition control unit 2a refers to the usage status history information corresponding to the user ID acquired in SA1 of FIG. 4 in the usage status history information of the usage status history information DB 3c updated in SD3. Then, the number of voice recognition commands selected so far is accumulated for each voice recognition command, and the priority of each voice recognition command is newly set so that the priority becomes higher in descending order of the cumulative result. Then, the newly set priority is compared with the priority already set in the voice recognition command table 3b in FIG. 2 corresponding to the user ID acquired in SA1 in FIG. Updates the already set priority with the newly set priority. By performing such updating, the priority of the voice recognition command in which the selection criterion classification = user adaptive type is always updated to the latest priority based on the selection history of the voice recognition command by the user. This completes the voice recognition command execution process.

(Processing-Voice recognition control processing-Same screen display processing)
Next, the same screen display process (SA6) in FIG. 4 will be described. 9 and 10 are flowcharts of the same screen display process. In this process, the voice recognition control unit 2a displays a standard command on the display 6 (SE1). This standard command can be selected in the same way as before. The standard command to be displayed first is the standard command at the highest level among the predetermined voice recognition commands hierarchized by the hierarchical structure. To do.

  Next, the voice recognition control unit 2a displays the same screen display command on the display 6 (SE2). Specifically, the speech recognition control unit 2a displays the adaptive commands selected in SB3 of the adaptive matching process in FIG. 5 and sorted in SB4 as the same screen display command in parallel with the standard command displayed in SE1.

  Here, the position and size of the display area of the standard command and the same screen display command can be arbitrarily determined, but should be determined in consideration of the display number and display size of these standard commands and the same screen display command. Is preferred. For example, when displaying the same screen display command as the standard command with the same display number and display size as each other, as shown in FIG. 11, the display of the standard command menu 21 composed of a plurality of standard commands 20 is displayed. It is preferable that the display area of the same screen display command menu 23 composed of the area and the plurality of same screen display commands 22 be half of the entire display area of the display 6. When the display area is halved in this way, it may be a left and right half area as shown in FIG. 11, or a top and bottom half area. In the case of the left and right half areas, it is preferable to display the standard command 20 in the left half area and the same screen display command 22 in the right half area as shown in FIG. Alternatively, when the upper and lower half areas are used, it is preferable to display the standard command 20 in the upper half area and display the same screen display command 22 in the lower half area. However, when the number of displays of the same screen display command 22 is not the same as the number of displays of the standard command 20 (for example, when the number of displays of the standard command 20 is larger than the number of displays of the same screen display command 22), When the display size of the same screen display command 22 is not the same as the display size of 20 (for example, when the display size of the standard command 20 is larger than the display size of the same screen display command 22), The display area with the larger display size may be larger than the other display area. In the display area determined as described above, the standard command 20 and the same screen display command 22 may be displayed vertically as shown in FIG. 11 or may be displayed horizontally. When displaying the same screen display command 22 in a vertical arrangement, it is preferable to display the commands in order of decreasing priority from top to bottom, and when displaying the same screen display command 22 in a horizontal arrangement, the display is from left to right. In order, it is preferable to display so that the priority becomes lower.

  Next, the voice recognition control unit 2a determines whether or not the update flag of the data recording unit 3 is ON (SE3). If the update flag is ON (SE3, Yes), the update is performed. The existence flag is turned OFF (SE4), and the same screen display command displayed in SE2 is output in voice in order of priority (SE5). These SE3 to SE5 can be performed in substantially the same manner as SC2 to SC4 in the map overlay display process of FIG. 6 except that the voice recognition command is not the overlay display command but the same screen display command. The detailed explanation is omitted.

  Next, the voice recognition command specifying unit 2c determines whether or not the user has operated the voice recognition command within a predetermined time (SE6) as in the case of SC5 in FIG. It is determined whether or not the operated voice recognition command is an operation of a standard command that is not the lowest hierarchy command (SE7). If it is determined that the operation is not a standard command that is not the lowest hierarchy command (SE7, No), it is an operation of the same screen display command or a standard command of the lowest hierarchy command, and the operated same screen display command Alternatively, since the operation content can be specified by the standard command, the voice recognition command specifying unit 2c starts the voice recognition command execution process for the same screen display command or the standard command (SE8). Since this voice recognition command execution process is the same as the voice recognition command execution process of FIG. 8 described above, detailed description thereof is omitted.

  On the other hand, if it is determined that the operation is a standard command that is not the lowest hierarchical command (SE7, Yes), the operation content cannot be specified by the operated standard command, so that the voice recognition command execution processing has not yet started, The recognition control unit 2a updates the standard command displayed on the display 6 (SE9). That is, the voice recognition control unit 2a replaces the standard command currently displayed on the display 6 with the standard positioned at the level immediately below the level where the standard command is positioned (one level below). The command is displayed on the display 6.

  Further, the voice recognition control unit 2a selects the same screen display command to be displayed next (SE10). That is, the voice recognition control unit 2a specifies the voice recognition command positioned in the lowest hierarchy corresponding to the hierarchy in which the standard command displayed on the display 6 is positioned after the update of SE9. A predetermined number of the same screen display commands are selected from the voice recognition commands, as in SB3 of FIG. For example, in the hierarchical structure of the voice recognition commands illustrated in FIG. 12, the voice recognition commands “command 1”, “command 2”, “command 3”, “command 4”, “command 5” that are first positioned in the highest hierarchy. ”Is displayed on the display 6, if the voice recognition command operated in SE 6 of FIG. 9 is“ command 2 ”, in SE 9 of FIG. 10, the voice recognition command“ command 2 ”is The voice recognition commands “C” to “F” positioned in the next lower layer are displayed on the display 6. In SE10, a predetermined number of the same screen display commands are selected from the speech recognition commands “P” to “X” positioned in the lowest hierarchy corresponding to these speech recognition commands “C” to “F”. select. As described above, the same voice recognition command can be prevented from being repeatedly displayed on the display 6 by selecting the same screen display command in a layer different from the standard command. In addition, by selecting the same screen display command positioned at the lowest layer corresponding to the standard command displayed on the display 6, the voice recognition command corresponding to the voice recognition command selected by the user can be directly The voice recognition command that can specify the operation content can be displayed on the screen, and the instruction of the operation content by the voice recognition can be shortcutly performed.

  Thereafter, in FIG. 10, the speech recognition control unit 2a determines whether or not the standard command updated in SE9 and the same screen display command selected in SE10 overlap (SE11). That is, by repeating the selection of the standard command, the standard command updated in SE9 becomes the standard command positioned in the lowest hierarchy, and may overlap with the same screen display command selected in SE10. In FIG. If there is an overlap (SE11, Yes), the same screen display command that is duplicated is excluded and the same screen display command is selected again in the same manner as in SE10 (SE12). For example, the same screen display command having the next highest priority is selected after the same screen display command excluded as being duplicated.

  Then, the voice recognition control unit 2a replaces the same screen display command currently displayed on the display 6 with the same screen display command reselected in SE12, or when there is no overlap (SE11, No), SE10. The same screen display command selected in (1) is displayed on the display 6 (SE13), and the same screen display process is terminated.

(Processing-Voice recognition control process)
After the map superimposition display process of FIG. 6 or the same screen display process of FIGS. 9 and 10, the process returns to the voice recognition control process of FIG. 4, and the voice recognition control unit 2 a turns off the navigation device 1. It is determined whether the notification is received from the navigation device 1 or other external system side (SA7). If not received (SA7, No), the process proceeds to SA2 to repeat the adaptive matching process (SA2). If received (SA7, Yes), the speech recognition control process is terminated.

(effect)
As described above, according to the present embodiment, since the voice recognition command selected based on the predetermined condition for determining whether or not the user is likely to use is displayed on the display 6, for example, a plurality of voice recognitions are performed. When commands are hierarchized in a hierarchical structure, the user can use the voice recognition commands without having to lower the hierarchical voice recognition commands from the highest hierarchy to the lowest hierarchy. The operability of voice recognition commands can be improved.

  Further, when a map is displayed on the display 6, the voice recognition command selected based on a predetermined condition for determining whether or not the user is likely to use is superimposed on the map and displayed. Even in the middle, it is possible to visually notify the user of voice recognition commands that are currently available, and it is possible to improve convenience when the user who is looking at the map uses the voice recognition commands.

  In addition, since the voice recognition command displayed on the display 6 is output as a voice, it is possible to notify the user of a voice recognition command that can be used at present by voice, and convenience for the user during driving to use the voice recognition command. And safety can be improved.

  In addition, since the priority order among a plurality of voice recognition commands is determined and the voice recognition commands are output in the order of the determined priority order, the user can select the voice recognition commands that can be used from the highest priority order. For example, since a voice recognition command with a high priority can be used without waiting for a voice recognition command with a low priority to be output, it is convenient for a driving user to use the voice recognition command. Property can be further improved.

  In addition, since the voice recognition command is selected based on the usage status history information, it is possible to select a voice recognition command that matches the user's past usage status, and to improve the user's convenience by increasing the selection accuracy of the voice recognition command. Can be further improved.

[Modifications to Embodiments]
Although the embodiments of the present invention have been described above, the specific configuration and means of the present invention can be arbitrarily modified and improved within the scope of the technical idea of each invention described in the claims. Can do. Hereinafter, such a modification will be described.

(About problems to be solved and effects of the invention)
First, the problems to be solved by the invention and the effects of the invention are not limited to the above contents, and may vary depending on the implementation environment and details of the configuration of the invention. May be solved, or only some of the effects described above may be achieved.

(Configuration of navigation device)
Further, each of the electrical components described above is functionally conceptual and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each part is not limited to the one shown in the figure, and all or a part thereof may be functionally or physically distributed / integrated in arbitrary units according to various loads and usage conditions. Can be configured. For example, the voice recognition device may be mounted on the vehicle as a voice recognition control system configured by distributing a plurality of devices, or all or part of the control unit 2 and the data recording unit 3 may be distributed on the network. May be. For example, the voice recognition device may be incorporated in a mobile phone or a personal computer.

(Adaptive command selection target)
In the above-described embodiment, it has been described that the lowest hierarchy command is selected as a voice recognition command that is likely to be used by the user. However, instead of the lowest hierarchy command or in addition to the lowest hierarchy command, The middle-level command may be selected.

(Adaptive command selection conditions)
As a predetermined condition for determining whether or not the user is likely to use, any other display algorithm may be used instead of or in addition to the display algorithm shown in the voice recognition command table of FIG. May be used.

(Adaptive command priority determination conditions)
As priority determination conditions for determining the priorities among a plurality of voice recognition commands, other arbitrary ones may be used instead of or in addition to the priorities shown in the voice recognition command table of FIG. Priority may be used. For example, the priority set by the user may be used. In the example of FIG. 2, the priority is set for each user ID for the voice recognition command in which selection criterion classification = user adaptive type, but the same priority is set for all users. May be.

(Adaptive command display method)
In the above-described embodiment, the example in which the adaptive command is superimposed on the map or displayed on the same screen as the standard command has been described. However, the adaptive command may be displayed by other methods. For example, when a map is displayed on the display, when displaying an adaptive command, only the adaptive command is displayed instead of the map, or the map and the adaptive command are alternately displayed at a predetermined interval, The adaptive command may be displayed non-superimposed on different display areas. Further, when displaying the standard command and the adaptive command, the standard command and the adaptive command may be alternately displayed at a predetermined interval.
In addition to displaying each adaptive command individually, for example, a voice recognition command “other” is displayed on the display 6, and when the voice recognition command is selected, it is displayed on the display 6 at that time. Other adaptive commands that are not performed may be displayed on the display 6.

(About usage history information)
The contents of the usage status history information recorded in the data recording unit and the priority update method using the usage status history information can be arbitrarily changed. For example, if the vehicle starts running before selecting the hierarchical voice recognition command up to the lowest hierarchy command, or if further operations using the voice recognition command are restricted for safety, the description Record the usage history information to identify the situation (traveling route, time from when the vehicle stopped to travel, etc.) and the voice recognition command that was displayed when the operation was restricted Keep it. After that, when it is specified that the situation is the same as the situation, the lowest hierarchy command corresponding to the voice recognition command that was displayed when the operation was restricted in the past is displayed. Thus, the operation using the voice recognition command may be easily completed before the operation is restricted.

DESCRIPTION OF SYMBOLS 1 Navigation apparatus 2 Control part 2a Voice recognition control part 2b Voice detection part 2c Voice recognition command specific part 3 Data recording part 3a Map information DB
3b Voice recognition command table 3c Usage history information DB
4 Microphone 5 Touch panel 6 Display 7 Speaker 8 Current position detection processing unit 10 Map 11 High-speed schematic diagram 12 Overlay display command 13 Overlay display command menu 20 Standard command 21 Standard command menu 22 Same screen display command 23 Same screen display command menu

Claims (7)

A voice detection means for detecting a user's speech;
Voice recognition command specifying means for specifying a voice recognition command included in the user's uttered voice based on the user's uttered voice detected by the voice detecting means;
Voice recognition control means for executing predetermined control corresponding to the voice recognition command specified by the voice recognition command specifying means, and selecting and displaying the voice recognition command on a display means;
The voice recognition control means causes the display means to display the voice recognition command selected based on a predetermined condition for determining whether or not the user is likely to use the voice recognition control means;
Voice recognition device. When a map is displayed on the display unit, the voice recognition control unit displays a voice recognition command selected based on a predetermined condition for determining whether or not the user is likely to use the map on the map. Superimposed and displayed on the display means,
The speech recognition apparatus according to claim 1. Voice output means for outputting voice to the user,
The voice recognition control means causes the voice output means to output a voice output for allowing the user to recognize the displayed voice recognition command when the voice recognition command is displayed on the display means.
The speech recognition apparatus according to claim 1 or 2. The voice recognition control means, when displaying a plurality of the voice recognition commands on the display means, determines the priority among the plurality of voice recognition commands based on a predetermined priority determination condition; In order of the determined priority order, the voice output means for outputting the voice output for allowing the user to recognize the voice recognition command,
The speech recognition apparatus according to claim 3. A usage history information storage unit for storing usage history information related to a usage history of a voice recognition command by a user;
The voice recognition control means includes
After displaying the voice recognition command on the display means, based on the usage status of the voice recognition command by the user, the usage status history information of the voice recognition command is stored in the usage status history information storage means,
When displaying a voice recognition command on the display means, a voice recognition command is selected based on the usage status history information stored in the usage status history information storage means.
The speech recognition device according to any one of claims 1 to 4. A voice detection step for detecting a user's speech;
A voice recognition command specifying step for specifying a voice recognition command included in the voice of the user based on the voice of the user detected in the voice detection step;
A voice recognition control step of performing predetermined control corresponding to the voice recognition command specified in the voice recognition command specifying step, and selecting and displaying the voice recognition command on a display unit;
In the voice recognition control step, the voice recognition command selected based on a predetermined condition for determining whether or not the user is likely to use is displayed on the display means.
Speech recognition method.   A speech recognition program for causing a computer to execute the method according to claim 6.

Images