JP2004109468A - Control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve a talk-back such that it can easily be judged whether facilities are ones that a user intends by a control system which places predetermined equipment in operation according to the name of the facilities as the recognition result of a voice that the user speaks. <P>SOLUTION: When the name of speech recognition is a facility name, the recognized facilities are specified (S50). Geographic information with which the presence position of the facilities can be grasped is acquired from a map data storage part (S60). Then geographic information is selected so that pieces of selected information are less than a given number (S80), "the name of facilities as a landmark" plus "a turn of phrase showing geographic relation" is added before the speech-recognized facility name, and a talk-back for the facility is performed (S90). Consequently, the user is able to judge whether facilities that the user intends are correctly recognized. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a control system for operating a predetermined device based on a name of a facility as a recognition result of a voice spoken by a user.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, an in-vehicle navigation system that sets an appropriate route (a destination route) from a current location to a destination, displays the destination route on a display together with a current position detected by GPS or the like, and provides route guidance has been developed. It is known and contributes to a smoother drive. At this time, the destination is input by the user himself. As this input method, for example, there is a method in which a destination is hierarchically searched from a menu display on a display, and a desired destination is designated if the desired destination is displayed. Manual input via is required. For this reason, in many cases, the destination can be input by voice in consideration of user's convenience. In particular, when the driver himself uses a so-called car navigation system, it is an effective input method because it does not involve any operation such as a switch operation or a screen gaze, and thus has high safety even when the vehicle is running. In the case of such a voice input, it is general to output a voice recognition result by voice (so-called talkback) so that the user can confirm whether or not voice recognition has been properly performed. It is. The user can confirm the content of the talkback, and if the word uttered by the user is erroneously recognized, a voice input can be performed again.
[0003]
Note that the above-described talkback is a general operation that is usually performed in a car navigation system capable of inputting voice, and corresponds to a known / public technology.
[0004]
[Problems to be solved by the invention]
However, the conventional voice talkback merely outputs the voice recognition result as it is, that is, talks back only the destination facility name or the like to which the user has input voice. Therefore, when a facility name is input by voice, there is a possibility that a destination not intended by the user is actually set even if the facility name is correctly recognized. For example, if a facility with the same name also exists in another place, if it is displayed on a map on the display, the user can know whether or not the destination is intended by taking the surrounding situation into account. However, in the case of talkback, such a surrounding situation is not known. Therefore, the user thinks that the intended facility is set as the destination because the name of the facility that he / she spoken is correctly spoken back. Nevertheless, there is a possibility that a facility of another location having the same name is actually set as a destination, and a route meaningless to the user is set.
[0005]
In addition, such a situation is not limited to route guidance.For example, when a map including the facility is displayed by designating the facility, detailed information such as business hours and charges at a facility such as a store or a parking lot is provided. The same applies to the case of notifying.
[0006]
Therefore, the present invention provides a control system that operates a predetermined device based on the name of a facility as a result of recognition of a voice uttered by a user, and a talkback that can easily determine whether the facility is intended by the user. The purpose is to realize.
[0007]
Means for Solving the Problems and Effects of the Invention
The control system of the present invention inputs a voice uttered by a user through a voice input unit, and operates a predetermined device based on the name of the facility as a result of recognition by the voice recognition unit. For example, a facility is set as a destination for route guidance as described in claim 7, a map including the facility is displayed as described in claim 8, and information on the facility is notified as described in claim 9. And so on. However, simply talking back the facility name as in the conventional method may not match the facility recognized by the control system with the facility recognized by the user.
[0008]
Therefore, according to the control system of the present invention, geographical information capable of grasping the location of the recognized facility is acquired using the map data, and the geographical information is talked with the name of the facility recognized by the voice recognition means. Back. Based on the geographical information around the facility talked back in this way, the user can determine whether or not the facility intended by the user is correctly recognized.
[0009]
Here, the geographical information to be talked back may be, for example, the address of the recognized facility, as described in claim 2. This is effective when the user knows the address where the facility is located.
However, the user does not always know the address where the facility is located. Even if you have been to the facility in the past, you may not always remember the address. Rather, they often remember the situation around the facility but not the address. Therefore, it is also an effective measure to talk back information indicating a geographical relationship with a predetermined landmark facility existing around the recognized facility, for example, as described in claim 3. For example, facilities using relatively large buildings such as schools, department stores, and stadiums are likely to be landmarks. Of course, not only the size, but also a relatively small facility such as a police box, convenience store, gas station, can be a landmark. And intersections can also be effective landmarks. An intersection name is attached to a traffic light at an intersection, and it is considered that the name is likely to remain in the user's memory. This is especially true when driving a car. Therefore, when talking back a geographical relationship such as "intersection angle", the user can easily understand whether the relevant intersection is near the facility intended by the user or is completely off-target. There is expected.
[0010]
Further, it is also effective to make the expression method of the information indicating the geographical relationship with the landmark facility different when the distance is relatively short and when the distance is relatively long. . For example, if it is relatively far, it is expressed as "near", and if it is relatively close, it is expressed as "-corner", "-before", "-next", "-north", and so on. With such a talkback, the user can easily imagine the positional relationship.
[0011]
In addition, when there are a plurality of landmark facilities, information indicating the geographical relationship with all the landmark facilities may be talked back. On the other hand, it is not preferable to talk back too much information. Therefore, for example, it is conceivable to talk back to about two. At this time, it is conceivable that the information to be narrowed down is based on the distance between the two as described in claim 5 for example. For example, two items may be selected in the order closest to the recognized facility, or the closest item and the farthest item may be selected. This is because if one of these two landmark facilities is memorized, the appropriateness of the recognized facility can be determined. Of course, three or more talkbacks may be performed depending on the control system applied.
[0012]
On the other hand, when the control system is mounted on a moving body, and when there are a plurality of landmark facilities in relation to the recognized facility, the function as a mark according to the property of the moving body is provided. It is conceivable to talk back information indicating a geographical relationship with only a predetermined number of landmark facilities based on the level of priority taking into account the priority. This is in view of the fact that the appropriate landmark facility depending on the application destination of the control system may be different. For example, when a control system is mounted on a vehicle as a moving body, there is a landmark facility that is easily recognizable to a user who is driving the vehicle. It talks back information.
[0013]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments to which the present invention is applied will be described with reference to the drawings. It is needless to say that the embodiments of the present invention are not limited to the following examples, and can take various forms as long as they belong to the technical scope of the present invention.
[0014]
Since the present embodiment is applied to an in-vehicle navigation system, a schematic configuration of the navigation system 1 will be described with reference to FIG. The navigation system 1 includes a position detection unit 11, a map data storage unit 12, a switch information input unit 13, a display unit 14, a memory unit 15, a voice output unit 16, a voice input unit 17, a control unit 20, and the like.
[0015]
The position detection unit 11 includes a well-known gyroscope, a distance sensor, a GPS (Global Positioning System) GPS receiver that detects the position of the vehicle based on radio waves from satellites, and the like. Since each of these sensors and the like has an error having a different property, the sensors and the like are configured to be used while being interpolated by a plurality of sensors. Note that, depending on the accuracy, a part of the above-described components may be used, and a geomagnetic sensor, a steering rotation sensor, a wheel sensor for each rolling wheel, or the like may be used.
[0016]
The map data storage unit 12 corresponds to a map data storage unit and stores various data including map data and so-called map matching data for improving the accuracy of position detection. The road data forms a map by connecting a plurality of nodes such as intersections with links, and for each link, a unique number (link ID) for specifying the link and a link indicating the length of the link It stores link information including data of length, x and y coordinates of the start and end of the link, road width of the link, and road type (indicating road information such as a toll road). In addition to the road data, the map data also includes display item data composed of characters and symbols indicating addresses, intersections, and facilities. The display item data has corresponding position information, and these display items (characters and symbols constituting the display items) can be superimposed and displayed at corresponding positions on the road map. The facilities referred to here are facilities that can be displayed on a map, and include a facility genre, a facility position when displayed on a map, and the like. As a storage medium, a CD-ROM, a DVD-ROM, a hard disk, or the like is generally used, but another medium such as a memory card may be used. Needless to say, since it is necessary to read data from these storage media, a CD player and a DVD player are also provided.
[0017]
The switch information input unit 13 includes, for example, a touch switch integrated with the display device of the display unit 14, a mechanical switch attached around the display device, or a talk switch described later, and is used for various inputs. You. The touch switch is composed of an infrared sensor that is arranged on the display screen endlessly and vertically. For example, when the infrared ray is interrupted by a finger or a touch pen, the interrupted position is detected as a two-dimensional coordinate value (X, Y). Is done. Various inputs may be performed by a remote controller (not shown), in which case a remote control sensor may be prepared.
[0018]
In addition, the display unit 14 displays, on its display screen, the vehicle current position mark input from the position detection unit 11, the map data input from the map data storage unit 12, and a guidance route to be further displayed on the map. The additional data such as the mark of the set point can be displayed in a superimposed manner.
[0019]
The memory unit 15 stores a program for executing various processes related to the navigation function, and is used as a work memory when executing the program. Further, it is also used for temporarily storing map data and the like obtained from the map data storage unit 12.
[0020]
Further, the audio output unit 16 is configured to notify a user (a driver or the like) of travel guidance by voice via a speaker. Therefore, the user can be guided by both the display on the display unit 14 and the audio output from the audio output unit 16.
[0021]
The voice input unit 17 corresponds to a voice input unit, and inputs voice emitted by the user via a microphone, converts the voice signal into digital data, and inputs the digital data to the control unit 20.
Next, the control unit 20 will be described. The control unit 20 is configured as a normal computer, and internally includes a well-known CPU, ROM, RAM, I / O, and a bus line connecting these components. As shown in FIG. 1, it is shown as a functional block. That is, the control unit 20 includes a map data acquisition unit 21, a map matching unit 22, a route calculation unit 23, a route guidance unit 24, a drawing unit 25, a screen control management unit 26, and a voice recognition / talkback control unit 27. .
[0022]
The map matching unit 22 specifies on which road the current position is located using the current position information detected by the position detection unit 11 and the road shape data of the map data stored in the map data storage unit 12. I do. Further, the user uses the switch information input unit 13 to input an instruction to display a desired map or the like, and further sets a destination.
[0023]
The route calculation unit 23 calculates information on the current position calculated by the map matching unit 22 and a route from the departure point set by the user to the destination. As a method for automatically setting the optimum route, a method such as the Dijkstra method is known. The route guidance unit 24 calculates points necessary for route guidance from the calculation result of the route calculation unit 23, road shape data stored in the map data, intersection position information, level crossing position information, and the like. And what kind of guidance is needed (for example, whether to turn right or left).
[0024]
The drawing unit 25 draws a map of the current position, a schematic diagram of an expressway, an enlarged view of an intersection near an intersection, and the like in the drawing memory unit 17 configured by a VRAM or the like in accordance with an instruction from the screen control management unit 26. At the time of this drawing, the display method of characters / symbols is determined according to the situation by the outline font technique. The drawn map and the like are displayed on the display unit 14 according to an instruction from the screen control management unit 26.
[0025]
On the other hand, the voice recognition / talkback control unit 27 corresponds to a voice recognition unit and a talkback control unit, and outputs a keyword (hereinafter referred to as a utterance) as a word spoken by a user from a voice signal input through the voice input unit 17. A voice recognition unit for recognizing and acquiring a keyword is also provided. The voice recognition unit includes a collation unit and a recognition dictionary unit. This recognition dictionary unit stores dictionary data for each of a plurality of keywords (recognition target vocabulary) that the user is supposed to speak. The collating unit performs collation (recognition) using the voice data input from the voice input unit 17 and the dictionary data of the recognition dictionary unit. Then, the recognition result is talked back. This talkback controls the audio output unit 16 to output the recognized result by audio. After that, a process according to the instruction from the user is performed. When the recognition result is a facility name, not only the facility name is talked back, but also geographical information that can grasp the location of the facility is talked back, which will be described later.
[0026]
In this embodiment, when the user presses a talk switch (not shown), which is one of the switches included in the switch information input unit 13, a voice input can be performed thereafter. Since it is assumed that no voice input is made even when the talk switch is pressed, if there is a silent section for a predetermined time or more after the talk switch is pressed and voice input becomes possible, voice input becomes impossible. Transition to state. Therefore, the voice input unit 17 monitors the timing at which the talk switch is pressed, and detects that the talk switch is pressed.
[0027]
The map data obtaining unit 21 obtains the map data required by each processing unit from the map data storage unit 12 and supplies the map data to each processing unit. Each of the above processes is executed based on a program in the memory unit 15 and is also executed using a work memory in the memory unit 15.
[0028]
Next, a process related to voice recognition will be described.
First, it is determined whether or not the talk switch is turned on (pressed) based on information from the switch information input unit 13 (S10). If the talk switch is turned on (S10: YES), voice extraction is performed. Processing is performed (S20). In the voice extraction processing, the voice input unit 17 determines whether the voice section is a voice section or a noise section based on voice data input via a microphone, extracts data of the voice section, and performs voice recognition / talkback. This is a process of outputting to the control unit 27.
[0029]
Next, voice recognition processing is performed (S30), and it is determined whether or not the recognition result is a facility name (S40).
If it is not a facility name (S40: NO), a normal talkback is executed (S100). The normal talkback is to talk back only the recognition result, and the voice recognition / talkback control unit 27 controls the voice output unit 16 and outputs the recognition result from the speaker by voice to recognize the user. This is to confirm the result.
[0030]
Then, whether or not the recognition is correct is determined based on an instruction from the user (S110). Specifically, the determination is made based on a voice input from the microphone by the user. For example, if there is a voice input indicating a positive content such as "Yes", it can be determined that the recognition is correct, and if there is a voice input indicating a negative content such as "No" or "No", the recognition is incorrect. I can judge. If the recognition is incorrect (S110: NO), the process returns to S10 and the process is repeated. On the other hand, if a positive determination is made in S110, that is, if it is determined that the recognition is correct, the recognition result is determined by the voice recognition / talkback control unit 27 (S120), and a predetermined post-determination process is executed (S130). The post-confirmation process in this case is, for example, a process in which if the recognition result is a “menu screen”, a command instructing display of the menu screen is input and output to the screen control management unit 26. After the process of S130, the process returns to S110 and repeats the process.
[0031]
On the other hand, if an affirmative determination is made in S40, that is, if the recognized result is a facility name, the processes in S50 to S90 are executed. First, the recognized facility is specified (S50). Specifically, the position coordinates and the like where the facility exists are specified. Note that data other than the position coordinates may be used as long as the location of the facility can be specified. However, in this embodiment, since the map data includes the position coordinates of each facility, the data is specified by the position coordinates. And Then, based on the specified position coordinates and the like, geographical information from which the location of the facility can be grasped is acquired from the map data storage unit 12 (S60), and if a predetermined number or more of the acquired geographical information is present, (S70: YES), geographic information is selected so as to be less than the predetermined number (S80). On the other hand, if the geographic information is less than the predetermined number (S70: NO), no geographic information is selected. When the geographic information is thus selected (S80) or when the geographic information is less than the predetermined number (S70: NO), a talkback for the facility is executed (S90). Then, after the process of S90, the process proceeds to S110. If a positive determination is made in S110, that is, if it is determined that the recognition is correct, the recognition result is determined by the voice recognition / talkback control unit 27 (S120), and a predetermined post-determination process is executed (S130). For example, in the case of a destination setting scene for route setting, the facility thus recognized and determined is output to the route calculation unit 23 as a destination. In addition, an inquiry for confirmation, such as "Is XX set as a destination?" Also, for example, when a map including the facility is displayed by designating the facility, or when, for example, detailed information such as business hours and charges is notified in a facility such as a store or a parking lot, etc. The determined facility is output to the execution subject of each process.
[0032]
This is an outline of the process of talking back the facility name, and each process will be supplementarily described.
(1) Acquisition of geographic information in S60
Since this geographical information is information for indicating the location of the facility specified in S50, geographical information around the specified facility is obtained. In the present embodiment, first, a landmark facility is extracted in order to talk back information indicating a geographical relationship with a predetermined landmark facility existing around the facility. The landmark facility may be, for example, a relatively large building such as a school, department store, or stadium, a relatively small facility such as a police box, a convenience store, or a gas station, or an intersection. Intersection names are often attached to traffic lights at intersections, and can be a valid landmark when the user actually goes to the place.
[0033]
Several examples of the method for extracting the mark facility will be described.
{Circle around (1)} First, it is conceivable to extract, for example, those existing within a range of a radius Xm centered on the position of the facility specified in S50. Since at least one marker facility is to be extracted, several levels of radius Xm are prepared. For example, first search within a radius of 50 m, and if no marker facility is found, then expand to a radius of 100 m. If one or more are found, the search may be terminated there. Note that even if the range is excessively widened, it cannot substantially be a landmark, so the upper limit radius is determined. If the facility for the landmark is not found even in the prepared widest range, the address of the facility specified in S50 is used, and the address of the facility specified in S50 is changed to "around the city" or "around the town". No geographic information may be employed.
[0034]
{Circle around (2)} A method of “setting from surrounding map sections” is also conceivable. Since the map data is divided and stored, it is searched within the block. At this time, if only one section is set as the search range, it is not possible to search for a facility near the end of the section even if there is another facility nearby. For example, if the facility specified in S50 exists near the end of the map section, The map section adjacent to the end may be searched. Since the size of the section changes according to the amount of map data, the search range may change for each area such as the so-called city center, suburbs, and mountain villages.
[0035]
{Circle around (3)} Also, a method of “setting at the same administrative division (larger character level or municipal level)” is conceivable. Also in this case, as in the case of the above (2), when a specific facility exists near the end of the administrative division, the adjacent administrative division is also set as the search range.
(2) Selection of geographic information in S80
In S80, if the acquired geographic information is equal to or more than the predetermined number (S70: YES), the selection is made to be less than the predetermined number. The predetermined number may be, for example, about three. In this embodiment, since the present invention is applied to an in-vehicle navigation system, talkback is basically performed for a user (driver) driving the vehicle. Since it is not preferable to talk back too much information to the user while driving, it is preferable to talk back to about two, for example. Therefore, for example, if there are three or more pieces of geographic information in S70, it is conceivable to narrow down to two or less in S80.
[0036]
Then, some examples of a method for narrowing down to geographic information will be described.
{Circle around (1)} For example, it may be based on the distance from the facility specified in S50. Specifically, two are selected in the order of “closest”, or “the closest and farthest” is selected. If one of these two landmark facilities is also memorized, the appropriateness of the recognized facility can be determined.
[0037]
{Circle around (2)} It is also conceivable to make a selection based on the “type”. Specifically, in the case of the present embodiment, since the navigation system is mounted on a vehicle, it is easy for the driver to recognize the navigation system, and the priorities of generally well-known facilities are set high, and such facilities are identified. That is, the priority is selected. For example, city halls, stations, stadiums, schools, etc. are effective as priority facilities. Also, intersections are considered to be very effective in that they can be checked from a car and that they can be found all over the country. In this case, it is possible to employ a plurality of intersections as landmark facilities, but in general, it is considered that it is often preferable to adopt another type of facility as a landmark facility, For example, when one intersection is selected, it is preferable to select the above-described city hall, station, stadium, school, or the like as a facility other than the intersection.
[0038]
(3) In addition to the viewpoints of (1) and (2) described above, it is conceivable to consider the wording when talking back. For example, the phrase "~ near-intersection angle" is preferable for the talked-back user to imagine the location of the facility, but those with the same phrase such as "~ near-near" represent the location of the facility. It is relatively unfavorable from the point of view of making it work. Therefore, it is conceivable to adopt a facility in a position where the same wording does not continue as much as possible.
(3) About talkback for facilities in S90
In the case of the normal talkback in S100, only the recognition result is talked back. However, in the facility talkback in S90, geographic information is added before the name of the facility specified in S50. Talk back. More specifically, talkback is performed on “name of landmark facility” + “phrase indicating the geographical relationship between the landmark facility and the facility specified in S50” + “name of facility specified in S50”.
[0039]
Here, for example, the following can be considered as a phrase indicating a geographical relationship.
{Circle around (1)} “Nearby”: When the distance to the target facility is far to some extent. For example, when the distance is a certain distance such as 100 m or more, the phrase “to near” is considered to match the user's sense.
[0040]
(2) "~ intersection angle": when the distance to the intersection is short. For example, if the distance is several tens of meters from the intersection, it is considered that “〜 intersection angle” can be used.
{Circle around (3)} “Before”: For example, if there is a destination across the main gate of an elementary school across a road, the phrase “△△ in front of elementary school” may be used. However, information on the front direction of the landmark facility is required, and exclusive data is required.
[0041]
{Circle around (4)} “Beside”: For example, if there is a target facility without a road beside the city hall, the phrase “□□ next to city hall” may be used.
(5) "-North", "-South", "-East", "-West": When the direction is clear, such expressions can be considered.
[0042]
Considering these (1) to (5), the relations such as (3), (4) <(5), (2) <(1) in order of decreasing distance from the target facility are as follows. Conceivable. This is because it is considered that it is preferable to use these expressions when it is a little distant compared to the case of “before” in (3) or the “to next” of (4). This is because the position can be specified “before” or “beside” regardless of the direction. Therefore, as to what kind of geographical relation wording is added to the landmark facility, for example, short-distance wording is preferentially selected, and any of the conditions (2) to (5) does not apply. If not, it is considered preferable to select "1."
[0043]
As a specific example of the talkback of “name of landmark facility” + “phrase indicating geographical relationship” + “name of specific facility performed in S50”, for example, “ Super @ Osone store? "," Is it a bookstore at the corner of Kariya station square? "
[0044]
When talking back in this way, it is possible to have dedicated geographical information, but in the present embodiment, information on indexes and display characters existing in map data is used. In current car navigation systems, index information and display character information are generally already stored as map data. In most cases, the Japanese syllabary search can be performed as a standard function, so that the index data includes reading data, and the read data can be used to generate a synthesized speech and talk back.
[0045]
Further, since the system of this embodiment has a voice recognition function, data for voice recognition may be used as it is. Since the voice recognition data is associated with the facility index data and the address index data, it can be easily realized by reusing them.
On the other hand, the display character information does not often include the reading data at present, and in that case, it is necessary to add the reading data. However, if such read data is added, it may be easier to provide dedicated data. When the dedicated data is created, it is considered that "coordinates", "information (type, etc.) that can be prioritized", and "talkback data (real voice or synthesized voice reading)" are required at a minimum.
[0046]
Further, as for the intersection, the voice data of the real voice is stored in the high-performance navigation even at present, so that it can be used.
As described above, in the navigation system 1 of the present embodiment, the user can, for example, voice-input a destination for route guidance with a facility name. However, simply talking back the facility name may not match the facility recognized by the navigation system 1 with the facility recognized by the user. If the route is set as it is despite such a recognition mismatch, the user would not have intended himself only after seeing the set route displayed on the display unit 14. When the user notices that a destination different from the destination has been set, the user has to start over from the destination setting again.
[0047]
Therefore, according to the navigation system 1 of the present embodiment, as the talkback for the facility, the name of the landmark facility + the wording indicating the geographical relationship is added before the voice-recognized facility name, and the talkback is performed. This allows the user to determine whether or not the intended facility is correctly recognized. For example, if a talkback "is a bookstore at the corner of Kariya station square in Kariya city?" Is made, the user thinks that the bookstore he is thinking of is actually another chain store in another area. You can notice that. For example, when there are a plurality of such chain stores, the user recognizes the facility intended by the user by inputting a voice together with the branch name or the like.
[0048]
When a plurality of candidate chain stores are extracted, for example, the talkback including “name of landmark facility” + “phrase indicating geographical relationship” is added in order, and the user, for example, “different” If the user inputs a voice such as "Yes", the user makes a positive determination in S110 of FIG. 2 and the recognition result is determined (S120). Is also good.
[0049]
[Another embodiment]
(1) In the above embodiment, the information indicating the geographical relationship with the predetermined landmark facility existing around the recognized facility is talked back. For example, the address of the recognized facility is talked back as geographical information. May be. For example, a talkback such as "Is this a bookstore at the address of △△ city ○ town □□□?" Such talkback is also effective when the user knows the address where the facility is located. However, the user does not always know the address where the facility is located. Even if you have been to the facility in the past, you may not always remember the address, so it may be more effective to talk back the geographical relationship with the landmark facility as in the above example. It is thought that there are many.
[0050]
(2) In the above embodiment, an example in which the present invention is applied to the in-vehicle navigation system 1 has been described. However, the present invention is not limited to this application, and may be applied to, for example, a terminal device carried by a person. . In the above-described embodiment, in the description of S80, it has been described that, when the acquired geographical information has a predetermined number or more, selection may be made based on the “type”. Then, in the case of application to an in-vehicle navigation system, the driver is easily recognized by the driver, and the priority of a generally well-known facility is increased. Therefore, when the present invention is applied to a terminal device carried by a human, the priority may be determined from the viewpoint that it is easily recognized by a walking person.
[0051]
Further, the present invention is not limited to the navigation system, but may be applied to any control system that operates a predetermined device based on the name of a facility as a recognition result of the voice uttered by the user.
(3) The function of realizing the above-described processing related to voice recognition / talkback in a computer system can be provided, for example, as a program activated on the computer system side. In the case of such a program, for example, it is possible to use the program by recording it on a computer-readable recording medium such as a flexible disk, a magneto-optical disk, a CD-ROM, and a hard disk, and loading and activating the computer system as needed. it can. Alternatively, the program may be recorded on a ROM or a backup RAM as a computer-readable recording medium, and the ROM or the backup RAM may be incorporated in a computer system.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating an overall configuration of a navigation system as an embodiment.
FIG. 2 is a flowchart showing processing related to speech recognition.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Navigation system, 11 ... Position detection part, 12 ... Map data storage part, 13 ... Switch information input part, 14 ... Display part, 15 ... Memory part, 16 ... Audio output part, 17 ... Audio input part, 20 ... Control Units 21 Map data acquisition unit 22 Map matching unit 23 Route calculation unit 24 Route guidance unit 25 Drawing unit 26 Screen control management unit 27 Voice recognition / talkback control unit

Claims (9)

Voice input means for inputting voice spoken by the user;
Voice recognition means for recognizing voice input via the voice input means,
A control system for operating predetermined equipment based on the name of the facility as a result of recognition by the voice recognition means,
Map data storage means for storing map data;
Using the map data stored in the map data storage means, obtains geographical information from which the location of the facility recognized by the voice recognition means can be grasped, and recognizes the geographical information by the voice recognition means. And a talkback means for talking back together with the name of the facility. The control system according to claim 1,
The control system, wherein the geographic information is an address of the recognized facility. The control system according to claim 1,
The control system, wherein the geographic information is information indicating a geographical relationship with a predetermined landmark facility existing around the recognized facility. The control system according to claim 3,
The information indicating the geographical relationship with the landmark facility is characterized in that the expression method is different depending on whether the distance between the recognized facility and the landmark facility is relatively short or long. system. The control system according to claim 3 or 4,
When there are a plurality of landmark facilities in relation to the recognized facility, the talkback means talks back information indicating a geographical relationship with only a predetermined number of landmark facilities based on the distance between the two. A control system, characterized in that: The control system according to claim 3 or 4,
The control system is mounted on a mobile object,
The talkback means, when there are a plurality of landmark facilities in relation to the recognized facility, a predetermined number based on a priority level taking into account a function as a landmark according to the property of the moving body. A control system characterized by talking back information indicating a geographical relationship with only a landmark facility. The control system according to any one of claims 1 to 6,
A control system for setting a facility recognized as a result of the voice recognition means as a destination for route guidance. The control system according to any one of claims 1 to 6,
A control system for displaying a map including facilities as a result of recognition by the voice recognition means. The control system according to any one of claims 1 to 6,
A control system for notifying information on facilities as a result of recognition by the voice recognition means.

Images