JP2001306089A - Voice processor and navigation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice processor and a navigation device which can improve operability. SOLUTION: When a voice recognition button 62 is pressed, a proposal presentation part 68 vocally presents a candidate for an operation indication that a user performs next via a loudspeaker 12. When the user vocally shows an affirmative or an inaffirmative view to the presented candidate for the operation indication, a voice recognition part 64 recognizes the voice to specify a corresponding character string. When a character string, outputted from the voice recognition part 64, is an affirmative such as 'Yes', an operation content setting part 70 sets the operation contents for an operation object device and output it according to the candidate for the operation indication corresponding to the view.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice processing device for performing various processes corresponding to input voice and a navigation device.

[0002]

2. Description of the Related Art Conventionally, there has been known a voice processing apparatus which performs voice recognition processing on a voice of an operation instruction uttered by a user and executes various operations based on the recognition content. Considering a case where various operation instructions are given to a navigation device using a voice processing device, a user selects one of several commands registered in advance for operation instructions and utters this command. There is a need to.

For example, when a non-interactive voice processing device is used, a user utters a plurality of commands according to predetermined rules. The input order of these commands is determined according to the processing of the voice processing device or the navigation device operated by using the voice recognition result, and the user can utter the necessary commands in a predetermined order. Various operations can be instructed to the navigation device simply by doing.

[0004] When an interactive voice processing device is used, the user simply issues necessary commands one after another in response to inquiries from the voice processing device, and performs various operation instructions to the navigation device. It can be performed. For example, the voice processing device inquires the user, "What to do?", "What to look for?", Etc., and the user is prepared in advance such as "Search", "Bank", etc. Just say a few words to tell the navigation device "look for a bank as the nearest facility"
Operation instruction to that effect.

[0005]

By the way, the above-mentioned conventional voice processing apparatus is convenient in that the operation by the user is not required, but on the other hand, a plurality of commands and the like to be voice input are required. There is a problem that it is difficult to operate until the user gets used to it because he needs to learn the information. For example, in the above-described non-interactive voice processing device, it is necessary to memorize various commands and the like to be input, including the order in which they are input, and it is necessary to look at an operation manual book until one gets used to it. Also, in an interactive voice processing device, it is not necessary to remember the order of commands and the like to be input by voice. However, in response to an inquiry from the voice processing device, what commands and the like can be voice input can be used. Each user needs to remember. For example, in response to the query "What do you want to do?", If the selection candidates "Search", "Information", and "Destination" are prepared in advance, the user can select the correct command from these three commands. Must be selected and input by voice, and it is not allowed to make other commands without permission and input by voice. Therefore,
A user who does not remember the command or the like which is a candidate for voice input must eventually check the operation manual, which is not easy to operate.

The present invention has been made in view of the above points, and an object of the present invention is to provide a voice processing device and a navigation device which can improve operability.

[0007]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, a voice processing apparatus of the present invention proposes an operation instruction to be performed next by a user by a suggestion presenting means, and responds to the presented suggestion. , When a positive or negative opinion by a user is given by voice, the voice is subjected to voice recognition processing by voice recognition processing means, and the obtained recognition result indicates a positive opinion. In some cases,
Based on the proposal corresponding to this opinion, the operation content for the operation target device is set by the operation content setting means.
In this way, the voice processing apparatus proposes the next operation instruction, and the user responds with “Yeah”,
The operation content can be set simply by saying simple words such as "yes", "different", and "no" to convey a positive or negative opinion. It is not necessary to memorize the command or the like, and the operability can be improved.

Further, it is desirable to further comprise learning means for learning the operation content set by the operation content setting means. By learning the operation content set by the operation content setting means, the user's personality, such that the operation content that is more frequently used by the user is given priority in the next proposal of an operation instruction. Can make a proposal that reflects

[0009] In addition, it is desirable that the above-mentioned suggestion presenting means make the following suggestion with a wider range of presented contents when the recognition result by the speech recognition processing means indicates a negative opinion. When a negative opinion is obtained, by making a proposal with a wider range of content to be presented, it is possible to increase the possibility of proposing an operation content that meets the user's intention.

Further, in the navigation device of the present invention, when a predetermined navigation operation is performed by the navigation processing means, a suggestion is provided by the proposal presenting means to the next operation instruction by the user. When a user's positive or negative opinion is indicated by voice, the voice is subjected to voice recognition processing by the voice recognition processing means, and the obtained recognition result indicates a positive opinion. In such a case, the operation contents for the navigation processing means are set by the operation contents setting means based on the proposal corresponding to this opinion. Then, the above-mentioned suggestion presenting means makes the following proposal in which the range of the presented content is widened when the recognition result by the speech recognition processing means indicates a negative opinion.

[0011] Generally, in a car navigation system, an operation instruction is given by voice input.
Although the operation by the user is not required, the operability can be improved. However, in the present invention, the navigation device proposes the next operation instruction. By simply saying simple words such as "Yes,""No," or "No," and giving a positive or negative opinion, you can set the operation details. There is no need to learn commands and the like, and operability can be further improved. Further, each time a negative opinion is obtained, by making a proposal with a wider range of content to be presented, it is possible to increase the possibility of proposing an operation content that meets the user's intention.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A navigation device according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration of an on-vehicle navigation device according to an embodiment to which the present invention is applied. The navigation device shown in FIG. 1 includes a navigation controller 1 for controlling the whole, a DVD 2, a disk reader 3, an operation unit 4, a GPS receiver 5, an autonomous navigation sensor 6, a beacon transceiver 7, an FM multiplex broadcast receiver 8, a display. It is configured to include a device 9, an audio processing device 10, a memory 11, and a speaker 12.

The navigation controller 1 controls the operation of the entire navigation device. The detailed configuration of the navigation controller 1 will be described later. The DVD 2 records various map data necessary for map display, route search, and the like. The disk reader 3 is 1
One or a plurality of DVDs 2 can be loaded.
The map data is read from D2. The loaded disk is not necessarily a DVD, but may be a CD. Alternatively, both a DVD and a CD may be selectively loaded.

The operation unit 4 includes a search key for giving a route search instruction, a route guidance mode key used for setting a route guidance mode, a destination input key, left / right / up / down cursor keys, a map reduction / enlargement key, and a display screen. Various operation keys such as a setting key for fixing an item at the upper cursor position are provided, and a signal corresponding to an operation state of the key is transmitted to the navigation controller 1. In addition, this operation part 4
It may be realized by a remote control unit.

The GPS receiver 5 receives radio waves transmitted from a plurality of GPS satellites and performs three-dimensional positioning processing or two-dimensional positioning processing to calculate the absolute position and direction of the vehicle (vehicle direction is the current Is calculated based on the own vehicle position and the own vehicle position one sampling time ΔT before), and these are output together with the positioning time. The autonomous navigation sensor 6 includes an angle sensor such as a vibrating gyroscope that detects the vehicle rotation angle as a relative direction, and a distance sensor that outputs one pulse every predetermined traveling distance. Detect orientation.

The beacon transceiver 7 performs bidirectional communication mainly with radio wave beacon transceivers installed on highways via radio waves, and also mainly uses optical beacon transceivers installed on general roads. By performing two-way communication with the traffic information center (hereinafter referred to as “VIC
VICS traffic information sent from the “S center”. The FM multiplex broadcast receiver 8 receives VICS traffic information included in multiplexed data superimposed on general FM broadcast. The above-mentioned radio beacon, optical beacon and F
When compared with M multiplex broadcasting, there is no difference in that both can receive VICS traffic information and the content thereof, but in the case of FM multiplex broadcasting, traffic information can be obtained in a wider reception area.

The display device 9 displays map information around the own vehicle together with a departure point mark, a destination mark, etc., based on the drawing data output from the navigation controller 1, and displays a guidance route and VICS information on this map. Is displayed. The voice processing device 10 presents a candidate for an operation instruction to be given to the navigation device (hereinafter, this will be referred to as an “operation instruction candidate”), that is, a suggestion about the operation instruction via the speaker 12 by voice, and The voice obtained by the user in response to the given operation instruction candidate is acquired, and when a positive opinion is obtained, the content of the presented operation instruction is output to the navigation controller 1. The detailed configuration of the audio processing device 10 will be described later.

The memory 11 is realized by a detachable information storage medium such as various small flash memory cards and the like, and relates to various operation instructions given to the navigation apparatus and their use frequency. Is stored. As described above, by realizing the memory 11 by the detachable information storage medium, even when another navigation device is used,
There is an advantage that the user can perform an operation using his own database. Details of the contents of the database stored in the memory 11 will be described later.

The speaker 12 outputs a voice for presenting a candidate for an operation instruction based on a voice signal output from the voice processing device 10. Next, a detailed configuration of the navigation controller 1 will be described. The navigation controller 1 shown in FIG.
Buffer 2 for displaying a predetermined map image or the like on the
0, map read control unit 22, map drawing unit 24, VRAM2
6. The image synthesis unit 28, a vehicle position calculation unit 30, a route search processing unit 32, and a guidance route drawing for performing calculation of the vehicle position, map matching processing, route search processing, and route guidance processing and displaying the results. Unit 34, an input processing unit 40 for displaying various operation screens for the user, and transmitting operation instructions from the operation unit 4 or the voice processing device 10 to each unit, an operation screen generation unit 42, and a beacon transceiver 7 And the like, and a reception data buffer 44 for displaying VICS information received by the VICS information processing unit and the like, and a VICS information drawing unit 46.

The data buffer 20 is for temporarily storing map data read from the DVD 2 by the disk reading device 3. Map reading control unit 22
When the vehicle position is calculated by the vehicle position calculation unit 30, a request to read map data in a predetermined range including the vehicle position is sent to the disk reading device 3, and the map data necessary for map display is read from the DVD 2. Data buffer 2
Store to 0.

The map drawing section 24 creates map drawing data for displaying a map around the position of the vehicle based on the map data stored in the data buffer 20. The map drawing data created by the map drawing unit 24 is stored in the VRAM 26 and read out by the image combining unit 28. The image synthesizing unit 28 superimposes the map drawing data read from the VRAM 26 and the drawing data output from the operation screen generating unit 42 and performs image synthesis. The image combined by the image combining unit 28 is displayed on the screen of the display device 9.

The vehicle position calculation unit 30 calculates the position of the vehicle based on the data detected by the GPS receiver 5 and the autonomous navigation sensor 6, and when the calculated position is not on the road in the map data. Then, a map matching process for correcting the own vehicle position is performed. The route search processing unit 32 searches for a traveling route that connects a preset destination and departure place under predetermined conditions. For example, under various conditions such as the shortest distance and the shortest time, a guidance route with the lowest cost is set. As a typical route search method, the Dijkstra method and the horizontal search method are known. Guide route drawing unit 34
Draws the guidance route set by the route search processing unit 32 on the map image.

Next, a detailed configuration of the audio processing apparatus 10 will be described. FIG. 2 is a diagram illustrating a configuration of the audio processing device 10. As shown in the figure, the voice processing device 10 includes a microphone 60, a voice recognition button 62, a voice recognition unit 6
4, a speech recognition dictionary storage unit 66, a suggestion presentation unit 68, an operation content setting unit 70, and a learning processing unit 72.

The microphone 60 converts the voice uttered by the user into an electric signal. The voice recognition button 62 is a push button switch that is pressed when the user inputs various operation instructions to the navigation device by voice,
When the voice recognition button 62 is pressed, the operation of presenting the operation instruction candidate by the voice processing device 10 starts.

The voice recognition unit 64 analyzes a voice signal output from the microphone 60 and performs a predetermined voice recognition process to specify a character string corresponding to the voice uttered by the user, and The recognition result is output to the operation content setting unit 70 and the like. The voice recognition dictionary storage unit 66 stores a signal waveform corresponding to a standard voice as a voice recognition dictionary.

When the voice recognition button 62 is pressed, the suggestion presentation section 68 presents an operation instruction candidate (that is, a proposal for an operation instruction) to the user. Specifically, in the present embodiment, the operation instruction candidate is determined by a combination of a plurality of commands and a genre set for each command.
An operation instruction candidate is presented by selecting any command and genre and outputting a corresponding audio signal to the speaker 12. For example, if “search” is selected as a command and “convenience store” (hereinafter referred to as “convenience store”) is selected as a genre, the suggestion presentation unit 68 outputs a voice such as “search for a convenience store?” Output from the speaker 12. Details of the command and the genre will be described later.

When a positive opinion is obtained from the user with respect to the presentation of the operation instruction candidate by the suggestion presentation unit 68, the operation content setting unit 70 deletes the operation instruction candidate presented at that time. It is set as the operation content and output to the navigation controller 1. Specifically, in the present embodiment, the opinion on the operation instruction candidate is a simple word, for example, if it shows a positive opinion, if it shows a negative opinion, such as "Yes" or "Yes" If so, the user is informed in advance that the user can input “No” or “No”. Therefore, the operation content setting unit 70
When an operation instruction candidate is presented and a character string corresponding to a voice uttered by the user is output from the voice recognition unit 64 in response to the operation instruction candidate, the character string The user's opinion is determined by examining whether the word is a word or a negative word such as “no”.

The learning processing unit 72 checks the user's preference, that is, the frequency of use of each operation content, based on the content (operation content) of the operation instruction input to the input processing unit 40 in the navigation controller 1, and A predetermined learning process for updating the contents of the database stored in 11 is performed. Details of the operation of the learning processing unit 70 will be described later.

The above-mentioned suggestion presenting section 68 serves as a suggestion presenting means, the microphone 60, the voice recognition section 64, and the voice recognition dictionary storage section 66 serve as a voice recognition processing means, and the operation content setting section 70.
Respectively correspond to the operation content setting means. The learning processing unit 72 corresponds to the learning processing unit, and the navigation controller 1 corresponds to the navigation processing unit.

The navigation device according to the present embodiment has such a configuration. Next, the details of the above-described commands and genres and the contents of the learning processing performed by the learning processing section 72 regarding these will be described. As described above, in the present embodiment, when presenting an operation instruction candidate, the operation instruction candidate is set by combining a plurality of commands prepared in advance and a genre prepared for each command. .

FIG. 3 is a diagram showing an example of the command.
As shown in the figure, in the present embodiment, three commands “search”, “information”, and “periphery” are prepared.
Specifically, the command “search” is used when searching for various facilities such as a convenience store near the own vehicle position, and the command “information” is the VICS sent from the VICS center.
The command is used to display traffic information, and the command "around" is used to display the periphery such as the current location.

Each command is given a selection probability corresponding to the frequency at which the command is used. In the example shown in FIG. 3, the selection probabilities given to the three commands are “search” 70%, “information” 20%, and “surroundings” 10%. This selection probability is initially
For example, the user sets an appropriate value, or an appropriate value is set in advance as an initial value.

Then, the learning processing section 72 checks the contents of the operation input to the input processing section 40 in the navigation controller 1, and when each command shown in FIG. Update the selection probability to
The relationship between each command and the selection probability is stored in the memory 11.

FIG. 4 is a diagram showing an example of a genre set for each command. As shown in the figure, in the present embodiment, the genre corresponding to the command “search” is
Various facilities such as convenience stores, banks, restaurants, and so on are set. Similarly, as the genre corresponding to the command “information”, characters, simple figures, and traffic jam information are set. Here, "character" is level 1 of VICS information.
It corresponds to the "character display type", and when this genre is selected, the road traffic information is displayed by about 30 characters. Similarly, "simple figure" corresponds to level 2 "simple figure display type" of VICS information, and when this genre is selected, road traffic information is displayed by a simple road figure. Similarly, "congestion information" corresponds to level 3 "map display type" of VICS information, and when this genre is selected, congestion information and the like are overwritten and displayed on a map image. Is done.

As the genre corresponding to the command "around", a current location and a destination are set. The current location corresponds to the current position of the vehicle, and the destination corresponds to the destination when a route search is performed and a guidance route is set. Further, as shown in FIG. 4, a genre set for each command is also provided with a selection probability corresponding to the frequency at which the genre is used.
The selection probability is independently given for each of the genre corresponding to the command “search”, the genre corresponding to the command “information”, and the genre corresponding to the command “peripheral”. For example, if the command is "Search", the corresponding genres are convenience stores, banks, restaurants, ...
The sum of the values of the selection probabilities given to each of ... etc. is 100%
The selection probability is set so that

In the initial state, for example, the user sets an appropriate value for the selection probability given to each of these genres, or an appropriate value is set in advance as an initial value. Then, the learning processing unit 72 checks the operation content input to the input processing unit 40 in the navigation controller 1, and when each genre shown in FIG. 4 is used, the selection probability corresponding to each genre is determined. Is updated, and the relationship between each genre and the selection probability is stored in the memory 11.

As described above, in the present embodiment, the relationship between each command and the selection probability and the relationship between each genre and the selection probability are stored in the memory 1 as a database in advance.
1 and the value of the selection probability is updated by the learning processing unit 72. Next, a basic operation procedure of the learning processing unit 72 will be described with reference to a flowchart.

FIG. 5 is a flowchart showing the operation procedure of the learning processing section 72. The learning processing unit 72 determines whether or not various operation instructions have been input from the operation unit 4 or the voice processing device 10 to the input processing unit 40 in the navigation controller 1 (step 100). If the operation has been performed, it is determined whether or not the operation content corresponds to the predetermined command shown in FIG. 3 described above (step 101). If there is no operation instruction input, or if the operation content does not correspond to the predetermined command, the process returns to step 100 and the subsequent processing is repeated.

When the operation content corresponds to a predetermined command, that is, any one of the above-mentioned "search", "information", and "periphery", an affirmative determination is made in step 101, and the learning process is performed. The unit 72 specifies a command and a genre constituting the operation content (step 1).
02). For example, the operation content of “Find a convenience store existing around the position of the own vehicle” is the navigation controller 1.
Is input, "search" is specified as the command, and "convenience store" is specified as the genre.

Next, the learning processing section 72 updates the contents of the database stored in the memory 11. In particular,
The value of the selection probability assigned to each command is updated so that the value of the selection probability assigned to the command specified in step 102 becomes relatively high (step 103). Further, the learning processing unit 72 updates the value of the selection probability given to each genre so that the value of the selection probability given to the genre specified in step 102 becomes relatively high (step 104). After that, the learning processing unit 72 returns to Step 100 and repeats the subsequent processing.

In the present embodiment, for the genre corresponding to the command "search", the selection probability is updated in accordance with the frequency of use of the genre, and the user stops at a facility corresponding to each genre. Also, the selection probability is updated. By performing such processing, for example, for a user who frequently uses a convenience store, the probability that “convenience store” is selected as the genre corresponding to the case where the command “search” is selected is increased. And the user's preferences can be better reflected. Next, an operation of updating the selection probability performed in the learning processing unit 72 when the user stops at various facilities will be described.

FIG. 6 is a flowchart showing the operation procedure of the learning processing section 72 when the user stops at various facilities. The learning processing unit 72 acquires map data around the own vehicle position from the data buffer 20 in the navigation controller 1, acquires own vehicle position information from the vehicle position calculation unit 30, and adds the acquired map data and own vehicle position information to the acquired map data. Based on this, it is determined whether there are various facilities within a predetermined distance (for example, within 50 m) from the position of the vehicle (step 200).

For example, if facilities belonging to the genre "convenience store" or facilities belonging to the genre "bank" exist within a predetermined distance from the position of the own vehicle, an affirmative judgment is made in step 200, and then the learning processing section 72 determines whether a predetermined time (for example, 30 seconds) has elapsed (step 201). As described above, in the present embodiment, it is determined that the user has stopped at this facility when a predetermined time has elapsed while various facilities are present within a predetermined distance from the vehicle position. Until the predetermined time elapses,
A negative determination is made in step 201 and step 2
The processing after 00 is repeated.

If the predetermined time has elapsed, step 2
At 01, the learning processing unit 72 outputs a voice asking the user whether or not to perform a predetermined learning process for updating the selection probability of the genre to which the facility belongs, for the facility determined to have stopped ( Step 202).
Specifically, for example, if a facility belonging to the genre “convenience store” exists within a predetermined distance from the own vehicle position,
By requesting the suggestion presentation unit 68, the learning processing unit 72 outputs a voice such as “Where do you go to a convenience store?” From the speaker 12.

Next, the learning processing section 72 determines whether or not to perform a predetermined learning process based on an instruction given by the user (step 203). Specifically, in the present embodiment, when the user double-clicks on the voice recognition button 62 (the operation of pressing the button twice consecutively within a predetermined time), the learning processing unit 72 performs the predetermined learning processing. I do.

If the user instructs to perform the learning process, an affirmative determination is made in step 203, and the learning processing unit 72 corresponds to the genre to which the facility existing within a predetermined distance from the own vehicle position belongs. The selection probability of each genre is updated so that the selection probability becomes relatively high (step 204). Then, the above-described step 200
And the subsequent processing is repeated.

If the user does not give an instruction to perform the learning process, a negative determination is made in step 203, and the learning processing unit 72 sets the learning processing unit 72 to a predetermined time (for example, 5 minutes).
Is determined (step 20).
5). Then, the learning processing unit 72 returns to Step 203 until the predetermined time has elapsed, and repeats the subsequent processing. If the predetermined time has elapsed, the learning processing unit 72 performs the above-described step 20 without performing the learning process.
After returning to 0, the subsequent processing is repeated.

Next, a detailed description will be given of the operation procedure when the voice processing apparatus 10 presents operation instruction candidates, sets operation contents according to the opinion given by the user, and outputs the operation contents to the navigation controller 1. . FIG. 7 is a flowchart illustrating an operation procedure of the audio processing device 10. The suggestion presentation unit 68 determines whether or not the voice recognition button 62 has been pressed (step 300), and when pressed, reads the database from the memory 11 (step 301).
Next, the suggestion presentation unit 68 sets an operation instruction candidate based on the contents of the database, and presents the set operation instruction candidate to the user via the speaker 12 (step 302).

Specifically, in the present embodiment, the number of times the operation instruction candidates are presented is set in advance.
Assume that it is set to times. In this case, the proposal presentation unit 6
8 sets the number of times each command is selected in consideration of the selection probability. Various methods are conceivable for setting the number of times each command is selected in consideration of the selection probability. For example, in the present embodiment, a result obtained by multiplying the value obtained by converting the selection probability given to each command to a decimal number by “3”, which is the number of times the operation instruction candidate is proposed, is given as
0 for less than 0.5, 1 for 0.5 or more and less than 1.5, 2 for 1.5 or more and less than 2.5
The number of times each command is selected is set according to the rule that the number of times is 2.5 or more and 3 times in the case of 2.5 or more. According to this rule, in the example shown in FIG.
Is 70%, so that 0.7 × 3 = 2.
As a result, the command “search” is selected twice. Similarly, since the selection probability corresponding to the command “information” is 20%, 0.2 × 3 = 0.6, and the command “information” is selected once. Since the selection probability corresponding to the command “peripheral” is 10%, 0.1 ×
3 = 0.3, and the command “peripheral” is selected 0 times, that is, not selected.

When the number of times each command is selected is set in this way, the command with the highest selection probability is selected from among the genres corresponding to the selected command with priority given to the command with the highest number of times. When selected, the setting of the operation instruction candidate is performed. Specifically, in the example described above, the commands are selected in order of priority (1) “search”, (2) “search”, and (3) “information”.
When the first "search" is selected, the corresponding genre is selected. In the example shown in FIG. 4 described above,
“Convenience store” which is the genre with the highest selection probability among the genres corresponding to “search” is selected. As a result,
An operation instruction candidate “search for a convenience store” is set.
Similarly, when the second “search” is selected, “bank” which is the genre with the second highest selection probability among the genres corresponding to “search” is selected. As a result, “search for a bank” is selected. Is set. When the third "information" is selected, the "congestion information" which is the genre having the highest selection probability among the genres corresponding to the "information" is selected.
Is selected, and as a result, an operation instruction candidate “view traffic jam information” is set.

In the above-mentioned algorithm for setting the number of times each command is selected, if the selection probability corresponding to any two of the three commands is 50%, any one of the commands is selected. Is also selected twice, which may exceed the preset number of times of presenting the operation instruction candidate three times. In order to cope with such a case, each command must be On the other hand, a fixed priority may be set in advance.
Specifically, for example, a unique priority order of (1) “search”, (2) “information”, and (3) “periphery” may be set in advance for each command. As a result, for example, even when the number of times that the two commands “search” and “information” are selected is twice, the command “search” is preferentially selected twice, and the command “information” is Selected only once.

When the operation instruction candidates are presented according to the procedure described above, the voice recognition unit 64 determines whether or not the user's voice has been input to the microphone 60 (step 303). When an input is made, a predetermined voice recognition process for specifying a character string corresponding to the voice is performed (step 304).

Next, the operation content setting unit 70 acquires the character string obtained as the recognition result from the voice recognition unit 64, and determines whether or not a positive opinion has been given by the user based on the character string. Is determined (step 305). If the user's opinion is not positive, a negative determination is made in step 305, and the operation content setting unit 70 notifies the suggestion presentation unit 68 of that. The proposal presenting unit 68 that has received the notification determines whether the number of times the operation instruction candidates have been presented exceeds a predetermined number (step 306). And
If the number of presentations is equal to or less than the predetermined number, the suggestion presentation unit 6
8 sets the next operation instruction candidate (step 307),
Returning to step 302 described above, the processing after the operation of presenting the operation instruction candidate is repeated.

When the opinion of the user is positive, an affirmative determination is made in step 305 described above, and the operation content setting unit 70 sets the operation content corresponding to the presented operation instruction candidate, and performs navigation. Output to controller 1 (step 308).

The learning processing unit 72 performs a predetermined learning process according to the above-described procedure according to the operation content set and output by the operation content setting unit 70, and updates the database stored in the memory 11. (Step 30
9). Thereafter, the process returns to step 300, and the processing after the determination as to whether or not the voice recognition button 62 has been pressed is repeated.

If the number of presentations exceeds the predetermined number, an affirmative determination is made in step 306 described above, and the voice input device 10 shifts to a conventional interactive operation input operation of an operation instruction (step 310). . Specifically, for example, the proposal presenting unit 68
? "And the like, and then the sound processing device 10
Is in a standby state until a user gives an operation instruction by voice. Then, when a predetermined command or the like is uttered by the user and collected by the microphone 60, a voice recognition process is performed by the voice recognition unit 64 to specify a corresponding character string, and the character string corresponding to the specified character string is determined. The operation content is set by the operation content setting unit 70 and output to the navigation controller 1.

Next, an example of a dialog performed between the user and the voice processing device 10 included in the navigation device of the present embodiment will be described. In the description of the dialogue example, the user is indicated by “U” (User), and the voice processing device 10 is indicated by “A”.
(Agent). (Specific Example 1) A: “Will you find a convenience store?” U: “Yeah” In this example, the operation instruction “Find the nearest convenience store” that is used most frequently by the user is used as an operation instruction candidate by the voice processing device 10. Presented, and for this presentation,
This shows a case where a positive opinion is obtained from the user. As described above, since the operation instruction candidate selected according to the frequency of use is presented from the voice processing device 10, when the user wants to give the operation instruction, the user can make an inquiry from the voice processing device 10 ( It is possible to easily give an operation instruction to the navigation device simply by uttering a simple word such as “Yes” in response to “presentation”.

(Specific Example 2) A: “Are you looking for a convenience store?” U: “No,” A: “Are you looking for a bank?” U: “Yeah” This example shows that “the nearest convenience store is the most frequently used by users. This indicates a case in which the operation instruction "Search for" is presented as the first operation instruction candidate, but a negative opinion is obtained from the user. Then, next, operation instruction candidates whose range is extended to those having the second highest use frequency are presented,
On the other hand, positive opinions have been obtained by users. As described above, by gradually expanding the range of the operation instruction candidates, the possibility of presenting the operation instruction candidates according to the intention of the user can be increased.

(Specific Example 3) A: “Are you looking for a convenience store?” U: “Difference” A: “Did you look for a bank?” U: “Difference” A: “Did you see traffic jam information?” U: “Difference” A: U: “Look for a restaurant” In this example, in accordance with the frequency of use by the user, the range is gradually increased from the most frequently used to the least frequently used, and three operation instruction candidates ( Finding the nearest convenience store, Finding the nearest bank, Looking at traffic congestion information) is presented, indicating that the user has obtained a negative opinion on any of the operation instruction candidates. Thus, the audio processing device 10
If none of the operation instruction candidates presented from the side does not meet the user's intention, the voice processing device 10 performs the voice output such as "Do you want to do something?" The operation shifts to a conventional interactive operation instruction input operation of giving an operation instruction from the user.

As described above, in the navigation device according to the present embodiment, the operation instruction candidate is presented to the user from the voice processing device 10 and the opinion on the operation instruction candidate is simply expressed as "Yes" or "No". Since the operation content given to the navigation device can be set only by inputting in simple words, it is not necessary to memorize a plurality of commands or the like in order to set the operation content, and the operability can be improved.

Further, since a predetermined learning process for updating the selection probability in accordance with the frequency of use of the command or the genre is performed, the next operation instruction is proposed for the operation content which is more frequently used by the user. Since the priority is given to the user, there is also an advantage that a proposal reflecting the personality of the user can be made.

The present invention is not limited to the above embodiment, and various modifications can be made within the scope of the present invention. For example, in the embodiment described above,
As shown in steps 200 to 201 in FIG. 6, when it is determined whether or not a user has stopped at a facility corresponding to each genre, facilities corresponding to each genre exist within a predetermined distance from the own vehicle position. Is determined to have continued for a predetermined time or more, but other determination conditions may be used, or a plurality of determination conditions may be used in combination.

Specifically, for example, in a state where a predetermined facility exists within a predetermined distance from the position of the own vehicle, the engine is turned off, and it is determined that the engine has been turned on again, and it is determined that the vehicle has stopped at the facility. Good. In this case,
When the engine is turned on again, it is sufficient to ask the user whether or not to perform a predetermined learning process.

Similarly, it may be determined whether or not the user has stopped at the facility based on the operation states of the turn signals and the hazard lamps of the vehicle. For example, in a state where a predetermined facility is present within a predetermined distance from the position of the own vehicle, when it is detected that a turn signal is emitted or a hazard lamp is turned on, the user determines whether to perform a predetermined learning process. You can contact.

Further, it may be detected that the vehicle has changed the course at a place other than the intersection, and it may be determined that the user has stopped at the facility. In this case, the road information and the intersection information included in the map data stored in the data buffer 20 in the navigation controller 1 are obtained, and the own vehicle position and the traveling direction are obtained from the vehicle position calculation unit 30. It is sufficient to check whether or not the course has been changed in a place other than. Then, when a course change other than at an intersection is detected, it is sufficient to ask the user whether or not to perform a predetermined learning process.

As described above in step 203 of FIG. 6, in the present embodiment, an instruction as to whether or not to perform a predetermined learning process is made by double-clicking the speech recognition button 62. However, the present invention is not limited to this, and may be performed by voice input.
Further, the instruction from the user may be automatically performed without turning. In the case of voice input, the learning processing unit 72
May obtain a character string output from the voice recognition unit 64 in response to the voice of the user, and perform the learning process when the character string is a positive word such as “Yeah”. .

Further, in the above-described embodiment, the number of operation instruction candidate presentations by the suggestion presentation unit 68 is set to three, but the present invention is not limited to this, and this number can be arbitrarily set by the user. May do it. Further, in the above-described embodiment, the navigation device according to the embodiment including the voice processing device 10 to which the present invention is applied has been described in detail. However, the present invention is not limited thereto. The voice processing device 10 may give an operation instruction to various other devices such as an audio device.

[0068]

As described above, according to the present invention, an operation instruction to be performed next is proposed, and in response to this, the user selects "Yes", "Yes", "No", "No", etc. Just say a simple word to tell a positive or negative opinion and set what you want to do.
There is no need to memorize a plurality of commands or the like in order to set the operation content, and operability can be improved.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration of a vehicle-mounted navigation device according to an embodiment.

FIG. 2 is a diagram illustrating a configuration of an audio processing device.

FIG. 3 is a diagram illustrating an example of a command.

FIG. 4 is a diagram showing an example of a genre set for each command.

FIG. 5 is a flowchart showing an operation procedure of a learning processing unit.

FIG. 6 is a flowchart showing an operation procedure of a learning processing unit when stopping at various facilities.

FIG. 7 is a flowchart showing an operation procedure of the audio processing device.

[Explanation of symbols]

 Reference Signs List 1 navigation controller 4 operation unit 5 GPS receiver 6 autonomous navigation sensor 9 display device 10 sound processing device 11 memory 12 speaker 20 data buffer 24 map drawing unit 30 vehicle position calculation unit 40 input processing unit 46 VICS information drawing unit 60 microphone 62 sound Recognition button 64 Speech recognition unit 66 Speech recognition dictionary storage unit 68 Suggestion presentation unit 70 Operation setting unit 72 Learning processing unit

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06N 3/00 550 G08G 1/0969 9A001 G08G 1/0969 G10L 3/00 571T G10L 13/00 Q 15/00 551Q 15/28 F term (reference) 2F029 AA02 AB09 AC02 AC08 AC14 AC18 AD04 AD07 5D015 KK01 LL11 5D045 AB17 5E501 AA22 AC03 BA05 CA04 CA06 CB15 DA17 EA21 FA05 FA14 FA32 FA45 5H180 BB04 BB13 BB20 FF04 FF20 FF20 FF20 FF20 FF20H

Claims (4)

[Claims] 1. A proposal presenting means for making a proposal for an operation instruction to be performed next by a user, and a positive or negative opinion by the user is shown by voice with respect to the proposal presented by the proposal presenting means. And a voice recognition processing means for performing a voice recognition process on the voice, based on the suggestion corresponding to the opinion, when the recognition result by the voice recognition processing means indicates a positive opinion. And an operation content setting means for setting the operation content for the operation target device. 2. The voice processing device according to claim 1, further comprising a learning unit that learns the operation content set by the operation content setting unit. 3. The proposal according to claim 1, wherein the suggestion presenting means, when the recognition result by the speech recognition processing means indicates a negative opinion, proposes a next proposal having a wider range of presentation contents. An audio processing device characterized by performing. 4. A navigation processing means for performing a predetermined navigation operation, a suggestion presentation means for making a proposal for an operation instruction to be performed next by the user, and an affirmation by the user for the proposal presented by the proposal presentation means. A voice recognition processing unit that performs voice recognition processing on the voice when a positive or negative opinion is indicated by the voice, and a recognition result by the voice recognition processing unit indicates a positive opinion Further comprising: operation content setting means for setting an operation content for the navigation processing means based on the proposal corresponding to this opinion, wherein the suggestion presentation means has a negative recognition result by the voice recognition processing means. If it is an opinion,
A navigation device characterized by making the following proposal with a wider range of contents to be presented.