JPH1138994A - Voice recognition system and navigation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a load on a user that the user has to input voice from a high-ranking hierarchy even if it is more natural to omit the high-ranking hierarchy when an object of the voice-input to a voice recognition system is of a hierarchical structure, and improve the use convinience of a voice recognition system. SOLUTION: For example, when it is desired to input by voice, 'Aichi pref., Kariya-shi Showa town' but it is mis-recognized as 'Aichi pref., Kariya-shi Shoei town', 'Aichi-pref.' or 'Aichi-pref., Kariya-shi' composing a higher hieralrchical structure than 'town' level is set as a high-ranking hierarchical list which can be omitted, and a vertex to start collating can be down even to the city (shi) level or the town level (S40). For this reason, only the second voice-input of 'Shouwatown' can be understood as an input with the part of 'Aichi pref., Kariya-shi' omitted, and a voice recognition processing is executed according to the inputted voice. Namely, it is enough for the user to voice-input only 'Shouwatown' corresponding to the mis-recognized part.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a speech recognition apparatus which is effective for, for example, enabling a destination setting or the like in a navigation system to be input by speech, and a navigation system having the speech recognition apparatus.

[0002]

2. Description of the Related Art Conventionally, there has been proposed a speech recognition apparatus which compares an inputted speech with a plurality of pattern candidates to be compared which are stored in advance and determines a speech having a high degree of coincidence as a recognition result. It has already been put into practical use, and is used, for example, for a user to input a destination name to be set in a navigation system by voice. In particular, when the driver uses the in-vehicle navigation system, voice input does not involve button operation or screen gaze, so that it is effective because the safety is high even when the vehicle is running.

In order to satisfy such a function, it is necessary to easily specify a sufficiently detailed point. Specifically, it is necessary to be able to input not only the level of the prefecture and the city but also the level of the name of the town under the city and the level of the capital in the town and village. Furthermore, if the user wants to set, for example, "Showa-cho, Kariya-shi, Aichi", it must be pronounced separately for each level, such as "Aichi-ken", "Kariya-shi", "Showa-cho". Then, since it is troublesome for the user, it is preferable that input (batch input) can be performed continuously.

[0004] However, in the case of such a configuration corresponding to the batch input, a situation in which the usability is degraded for the user may be assumed. That is when the user has to re-enter the destination to be set. That is,
This is because the current recognition technology cannot always say that the recognition result is completely accurate, so that a single voice input does not always result in accurate recognition. For example, when the user inputs a voice as "Showa-cho, Kariya-shi, Aichi", for example, "Shoei-cho, Kariya-shi, Aichi"
May be mistakenly recognized. Then, in the case of such an erroneous recognition, it is necessary to input a voice again as "Showa-cho, Kariya city, Aichi prefecture".

However, considering conversations in daily life, in the case of such an erroneous recognition, it is necessary to correct only the erroneously recognized town name, instead of inputting a voice as "Showa-cho, Kariya-shi, Aichi Prefecture". Is natural. That is, it is considered preferable that the second time be able to input only "Showa-cho" by voice.

Although not caused by such an erroneous recognition, a vehicle equipped with, for example, an on-vehicle navigation system is traveling in, for example, Aichi Prefecture. When setting as a destination, it is more natural to input a voice as "Showa-cho, Kariya-shi" without omitting "Aichi-ken", instead of inputting a voice as "Showa-cho, Kariya-shi, Aichi".

[0007] Such a problem is not limited to the above-mentioned place names consisting of prefectures, municipalities, etc., but can be similarly applied as long as a plurality of words are similarly set as a hierarchical connection. . The present invention solves such a problem, and when the target of voice input to the voice recognition device has a hierarchical structure, even if it is more natural to omit the upper layer, voice input from the upper layer is not performed. The purpose is to reduce the burden on the user who must not do so and to improve the usability.

[0008]

Means for Solving the Problems and Effects of the Invention According to a first aspect of the present invention, there is provided a voice recognition device for inputting a voice, and a voice input device for inputting a voice. Means for comparing the input voice with a plurality of comparison target pattern candidates stored in advance in the dictionary means to obtain a recognition result having a high degree of coincidence, and a notifying means for notifying the recognition result by the recognition means And a post-confirmation processing means for executing a predetermined post-confirmation process assuming that the recognition result has been confirmed when a predetermined confirmation instruction is issued after the notification result is notified by the notification means. Wherein at least one of the plurality of comparison target pattern candidates stored in the dictionary means is:
A plurality of words are set as hierarchically connected, and the words or word groups constituting the upper layer of the comparison target pattern candidate in which the plurality of words are connected hierarchically are omitted in the recognition processing. In the case where the input speech is set, the recognition unit also temporarily considers the set upper-level constituent word or word group omitted as the comparison target pattern candidate. Characterized by performing a comparison against

[0009] According to the voice recognition device of the first aspect,
When the user inputs a voice through the voice input unit, the recognition unit compares the input voice with a plurality of comparison target pattern candidates stored in the dictionary unit in advance, and recognizes the one having a high degree of matching. And a notification unit notifies the recognition result. Then, when a predetermined confirmation instruction is given after the recognition result is notified, the post-confirmation processing means executes the predetermined post-confirmation processing assuming that the recognition result has been confirmed. Here, at least one of the plurality of comparison target pattern candidates stored in the dictionary means is set as a plurality of words connected hierarchically. Further, when a word or a group of words constituting an upper layer of the comparison target pattern candidate in which a plurality of words are hierarchically connected is set as an object to be omitted in the recognition processing, the recognizing unit sets the word. Also, the comparison with the input voice is executed after temporarily omitting the upper hierarchical constituent word or the word group as a candidate for the comparison target pattern.

Therefore, even if the comparison target pattern candidate is other than the comparison target pattern candidate previously set in the dictionary means, the upper layer constituent word or word group of the comparison target pattern candidate set as the omission target in the recognition processing is omitted. Is also temporarily regarded as a comparison target pattern candidate and the comparison with the input voice is executed, so that a voice input omitting such a higher hierarchical constituent word or word group is also allowed.

In order to clarify the effect of this, two specific examples which are set as objects to be omitted in the recognition processing will be described. First, as a first specific example set as an omission target in the recognition processing, for example, as described in claim 2, within a predetermined period after notifying the recognition result, a predetermined category same as the previous recognition result is set. When a speech input belonging to is made, a word or a group of words constituting an upper layer of the previous recognition result can be considered. As a result, for example, even though the user has input by voice as "Showa-cho, Kariya-shi, Aichi Prefecture", the user has erroneously recognized the voice recognition device as "Shoei-cho, Kariya-shi, Aichi Prefecture". In this case, when the user inputs the voice again, instead of inputting the voice of "Showa-cho, Kariya-shi, Aichi", it is sufficient to input only the voice of "Showa-cho". In other words, the first voice input, "Showa-cho, Kariya-shi, Aichi" is a hierarchical connection of multiple words indicating different levels of place names, "Aichi-ken", "Kariya-shi", and "Showa-cho". Therefore, "Aichi Prefecture" or "Kariya City, Aichi Prefecture" which is a word or a group of words constituting the higher hierarchy is set as an omission target in the recognition processing. Therefore, even if the user only inputs a voice for "Showa-cho" for the second time, it is possible to perform comparison with the input voice on the assumption that the input is "Kariya-shi, Aichi". Of course, it is possible to cope with the situation where only Aichi Prefecture is omitted and the voice is input again as "Showa-cho in Kariya city".

As described above, when a device is erroneously recognized, only the erroneously recognized portion (the town name "Showa-cho" in the above example) is corrected because of habits such as conversation in daily life. When you think about it, it's very natural. Forcing special attention only when using a voice recognition device is not preferable in terms of usability. Therefore,
As in the speech recognition apparatus of the present invention, it is possible to cope with a very natural behavior in a daily conversation habit of correcting only the erroneous recognition part. The user's burden of having to enter the information can be reduced, and usability can be further improved.

As a second specific example set as an object to be omitted in the recognition process, for example, as described in claim 4, the voice recognition in a system using the voice recognition result by the voice recognition device is used. Input from a device other than the device is conceivable. More specifically, if a voice recognition device is used to input a destination or the like by voice in a car navigation system, for example, current location information from a navigation device body in the system may be considered. In other words, if a vehicle equipped with a navigation system is traveling in Aichi Prefecture, for example, and if you set `` Showa Town in Kariya City, Aichi Prefecture '' as the destination in the same Aichi Prefecture, `` Showa Town in Kariya City, Aichi Prefecture '' It is more natural to omit "Aichi Prefecture" and input "Showa-cho Kariya" instead of voice input. In other words, there is a situation in which the user thinks that designation from the city level, which is the lower hierarchy, is sufficient even if the user does not specify from Aichi Prefecture.

Even in such a case, the speech recognition apparatus of the present invention can cope with such a case. In other words, in the above example, if the vehicle is currently traveling in Aichi Prefecture, the upper hierarchy of Aichi Prefecture may be set as an omission target in the recognition processing. As described above, in a predetermined area in which the user is currently located, a place in the upper hierarchy for specifying the area is not dared to be specified, but is specified from the lower hierarchy. By being able to respond to the behavior, even if it is more natural to omit the upper layer, it is possible to reduce the user's burden of having to input voice from the upper layer, and to further improve the usability.

In the above-described speech recognition apparatus, the recognition means temporarily omits the set upper-layer constituent word or word group as a comparison target pattern candidate, Is executed, two examples of the configuration for executing the comparison are shown below.

First, as set forth in claim 5, a reference data storage means for temporarily storing, as reference data, an upper layer constituent word or word group set as an object to be omitted in the recognition processing. Wherein the recognition means omits words constituting the upper layer of the reference data stored in the reference data storage means from the plurality of comparison target pattern candidates stored in the dictionary means when performing the recognition processing. Is temporarily regarded as a comparison target pattern candidate, and then the comparison is performed with respect to the current input voice.

Second, when the previous recognition result is obtained by connecting a plurality of words hierarchically, the recognition result is temporarily stored as reference data. A reference data storage unit that stores the reference data stored in the reference data storage unit, and a second configuration in which words constituting the upper layer of the reference data are omitted from the corresponding comparison target pattern candidates stored in the dictionary unit. Dictionary control means for newly establishing a comparison target pattern candidate, wherein the dictionary control means uses the comparison target pattern candidate stored in the dictionary means in a state where the second comparison target pattern candidate is newly established, and recognizes the recognition means. Is a configuration for executing comparison with the current input voice.

Also, when a predetermined confirmation instruction is given after the notification of the recognition result, it is described that the recognition result has been determined and the process proceeds to the predetermined post-confirmation processing. For example, when used in a car navigation system, a process itself for setting a destination as a recognition result or a process for instructing a device that executes the destination setting process to set the destination may be considered. . In addition, the "predetermined confirmation instruction" after the notification of the recognition result is input by voice (for example, input by uttering "yes") or instructed by operating switches such as a confirmation button. It is possible.

The condition for determining the recognition result excluding the pattern to be compared is that a voice input is made again within a predetermined period after the notification of the recognition result. As the “inside”, it is conceivable that the predetermined confirmation instruction is issued after the notification of the recognition result. In other words, when the confirmation instruction is given and the process proceeds to the predetermined post-confirmation processing, it means that the recognition result is correct. Therefore, the first recognition processing for the next voice input is set to be omitted in the previous processing. Do not continue what you have set. In particular, when a word or a group of words constituting the higher hierarchy is set as an abbreviated target from the previous recognition result, it is based on erroneous recognition on the device side, and is set as a recognition process for the next speech input. This is because carrying over is basically inappropriate. However, in the case where the name of the prefecture is omitted based on the current location, it is conceivable that the object to be omitted is the same as a result.

Further, regarding the notification of the recognition result,
This may be performed by outputting the content of the recognition result as a voice from a predetermined voice generating device. When used for in-vehicle devices such as car navigation systems,
If the voice is output, the driver does not need to shift the viewpoint to the display device, which is advantageous in terms of further ensuring safe driving. However, the present invention is not limited to audio output, and the content of the recognition result may be displayed on a display device capable of displaying characters or symbols on a screen by displaying images of the characters or symbols. May be notified, or other notification methods may be employed. It was mentioned that voice output is advantageous when applied as an in-vehicle device, but of course there are also situations where the vehicle is not running, so if both the voice and the image are used, the driver can confirm with the display and with the voice Both are possible.

When the speech recognition device according to any one of claims 1 to 7 is used for a navigation system, it is conceivable that the speech recognition device is configured as shown in claim 8. That is, a voice recognition device according to any one of claims 1 to 7 and a navigation device are provided, and the voice input means of the voice recognition device needs to be specified at least when the navigation device performs a navigation process. The user is used to input an instruction of predetermined navigation-related data by voice, and the post-confirmation processing means includes:
It is configured to output the result of recognition by the recognition means to the navigation device. In this case, the “predetermined navigation processing-related data” is a typical destination, but other than that, it is necessary to specify the navigation processing, such as selecting conditions for a route search. Instructions are included. In this case, the navigation processing-related information as the recognition result is notified. As described above, when a part of the place name once input by voice is erroneously recognized, the erroneous lower hierarchical layer is recognized. The user's usability is improved, for example, the user only has to re-enter the password.

It should be noted that, in particular, in the point that the name of the prefecture where the vehicle is currently traveling can be omitted and input can be made.
Are preferred. That is, the voice recognition device according to claim 4 and a navigation device, wherein the voice input means of the voice recognition device is provided with at least predetermined place name-related data required to be specified when the navigation device performs a navigation process. Is used for the user to input the instruction by voice, and the post-confirmation processing means is a navigation system that outputs the result of recognition by the recognition means to the navigation device, wherein the navigation device detects the current location. If the current position detected by the current position detecting means and the current position detected by the current position detecting means can be represented as the comparison target pattern candidate in which a plurality of words are hierarchically connected, a word or a group of words constituting the upper hierarchy is recognized. And an omission target data output means for outputting to the voice recognition device as an omission target at the time of processing. That.

In this case, since the navigation device has a function of detecting the current position in order to exhibit its basic function, for example, the prefecture name of the place where the vehicle is currently traveling is set as an abbreviation target from the detected current position. be able to.
The application destination of the voice recognition device is not limited to the navigation system described above. For example, when the voice recognition device is used for an air conditioning system, it is conceivable to adjust the set temperature, select an air conditioning mode (cooling / heating / dry), or select a wind direction mode by voice input.

For example, as for the set temperature, there are instructions relating to the set temperature, such as "decrease the set temperature to 25 degrees" or "decrease the set temperature by 5 degrees". . If the present invention is applied, if the user inputs "set the temperature to 25 degrees" but erroneously recognizes "set the temperature to 22 degrees", re-inputting only "set the temperature to 25 degrees" Just type in,
After all, the usability of the user is improved. The same applies to the air conditioning mode and the wind direction mode.

In other words, “a candidate pattern for comparison set as a plurality of words hierarchically connected”
In the navigation system described above, not only the hierarchical relationship of prefectures, cities, and so on like place names, but also the adjustment target (for example, set temperature) in the air conditioning system described above is set to the upper hierarchy, and the adjustment content (for example, 25 degrees) is made. Can be a lower layer.

The above navigation system and air conditioning system are not limited to the case where they are used as in-vehicle equipment, but may be, for example, a portable navigation device or an indoor air conditioner. However, as described above, when used for in-vehicle equipment, the user may be a driver. In that case, driving itself is the most important. For other in-vehicle equipment, drive as much as possible. Preferably, there is no hindrance. Therefore, in the case of a voice recognition device on the premise of a navigation system or an air conditioning system as an in-vehicle device, there is a further advantage. Of course, from such a viewpoint, the present invention can be similarly applied to in-vehicle devices other than the navigation system and the air conditioning system. For example, a car audio device is effective. Also, considering a configuration in which opening and closing of the power window and adjustment of the mirror angle are instructed by voice, it is effective even in such a situation.

[0027]

FIG. 1 is a block diagram showing a schematic configuration of a car navigation system 2 to which a voice recognition device 30 according to an embodiment of the present invention is applied. The car navigation system 2 includes a position detector 4, a map data input device 6,
An operation switch group 8, a control circuit 10 connected thereto,
External memory 12 connected to control circuit 10, display device 1
4 and a remote control sensor 15 and a voice recognition device 30. The control circuit 10 is configured as a normal computer, and has a well-known CPU, ROM, R
A bus line for connecting AM, I / O, and these components is provided.

The position detector 4 includes a well-known geomagnetic sensor 16, a gyroscope 18, and a distance sensor 2.
0, and GP for Global Positioning System (GPS) that detects the position of the vehicle based on radio waves from satellites
It has an S receiver 22. These sensors, etc.
Each of 8, 20, and 22 has an error having a different property, and is configured to be used while being interpolated by a plurality of sensors. It should be noted that depending on the accuracy, a part of the above-described components may be used, and a rotation sensor for the steering wheel, a wheel sensor for each rolling wheel, or the like may be used.

The map data input device 6 is a device for inputting various data including so-called map matching data, map data and landmark data for improving the accuracy of position detection. As a medium, CD-R
Although OM is generally used, another medium such as a memory card may be used.

The display device 14 is a color display device. On the screen of the display device 14, a vehicle current position mark input from the position detector 4, map data input from the map data input device 6, and a map It is possible to superimpose and display additional data such as a guidance route displayed above and a mark of a set point described later.

The present car navigation system 2
Is input from the remote control sensor 15 via a remote control terminal (hereinafter referred to as a remote control) 15a or the position of the destination by the operation switches 8;
It also has a so-called route guidance function that automatically selects an optimal route from the current position to the destination and forms and displays a guidance route. As a technique for automatically setting the optimum route, a technique such as the Dijkstra method is known.
The operation switch group 8 is, for example, a touch switch integrated with the display device 14 or a mechanical switch, and is used for various inputs.

The voice recognition device 30 is used when the operation switch group 8 or the remote controller 15a is manually operated to indicate a destination or the like, and similarly when the user inputs voice. This is a device that allows the user to specify a destination or the like.

The voice recognition device 30 includes a voice recognition unit 31 as a "recognition unit", a dialogue control unit 32, a voice synthesis unit 33, a voice input unit 34, and a microphone 35 as a "voice input unit". , PTT (Push-To-Talk) switch 36
And a speaker 37.

The voice recognition unit 31 performs a recognition process on the voice data input from the voice input unit 34 according to an instruction from the dialog control unit 32, and returns the recognition result to the dialog control unit 32. That is, the voice data acquired from the voice input unit 34 is collated using the stored dictionary data, and a higher-level comparison target pattern having a higher degree of matching as compared with a plurality of comparison target pattern candidates is determined. Output to The recognition of the word sequence in the input voice is performed by the voice input unit 3
The sound data inputted from the step 4 are sequentially subjected to acoustic analysis to extract an acoustic feature (for example, cepstrum), and to obtain acoustic feature time-series data obtained by the acoustic analysis. Then, the time-series data is divided into several sections by a well-known DP matching method, and it is determined which word each section corresponds to stored in the dictionary data.

From the recognition result and the internal state managed by itself, the dialogue control unit 32 navigates, for example, to the control circuit 10 which instructs the voice synthesizer 33 to generate a response voice and executes the processing of the system itself. A process for notifying a destination necessary for the process and instructing to execute the setting process is executed. Such processing is post-confirmation processing. As a result, if this voice recognition device 30 is used, the operation switch group 8 or the remote controller 15a can be operated without manual operation.
By voice input, it is possible to instruct a destination to the navigation device.

The voice input unit 34 converts the surrounding voice captured by the microphone 35 into digital data and outputs the digital data to the voice recognition unit 31. In the present embodiment, the user presses the PTT switch 36 while pressing the microphone 3.
This is a usage method of inputting a voice through the terminal 5. Specifically, the voice input unit 34 determines whether or not the PTT switch 36 has been pressed. When the PTT switch 36 has been pressed, the voice input unit 34 executes voice input processing via the microphone 35. If not, the voice input process is not executed. Therefore, only the voice data input via the microphone 35 while the PTT switch 36 is pressed is output to the voice recognition unit 31.

Here, the speech recognition unit 31 and the dialogue control unit 32
Will be further described. FIG. 2 shows the voice recognition unit 31.
4 shows the configuration of the dialogue control unit 32 in more detail, and two configuration examples (A) and (B) will be described. First,
In the configuration shown in FIG. 2A, the voice recognition unit 31
1a and a dictionary unit 31b.
2 comprises a storage unit 32a, an input unit 32b, and a post-processing unit 32c. In the voice recognition unit 31, the matching unit 3
1a compares the voice data acquired from the voice input unit 34 with the dictionary data stored in the dictionary unit 31b. For details, first refer to the storage unit 32a of the dialog control unit 32. If there is data to be created, an upper layer list that can be omitted is created with reference to the data. Then, collation is performed after temporarily omitting the upper layer part that can be omitted from the dictionary data stored in the dictionary unit 31b as a comparison target pattern candidate. The details of this collation will be described later.

Then, the higher-order comparison target pattern which is compared with the plurality of comparison target pattern candidates by the collating unit 31a and whose degree of coincidence is high is output to the storage unit 32a of the dialogue control unit 32. By the way, the dictionary data stored in the dictionary unit 31b of the present embodiment includes not only the data of the vocabulary itself as a comparison target pattern candidate, but also a vocabulary serving as the comparison target pattern in which a plurality of words are hierarchically connected. If it is, the data indicating the hierarchical structure is also stored. More specifically, the syllable data constituting the vocabulary is divided into the respective sides (FIG. 3) of the tree structure as shown in FIG.
Are indicated by arrows (→)). In FIG. 3, a single circle (○) represents a vertex, and a double circle (◎) represents a received vertex, that is, a vertex for a word.
Then, the vertex indicated by the arrow A in FIG. 3 becomes the “root”, from which the word is completed by tracing syllables assigned to each side according to preorder traversal.
Here, "preceding forward scan" means that a root is visited, and then a subtree having a child as a root is sequentially scanned (this scan is also a forward forward scan). Here, “parent” means the immediately preceding vertex, “child” means the next vertex, and “sibling” means vertices having the same parent.

That is, in the specific example shown in FIG. 3, when the vertices (indicated by arrow A) which are the roots are sequentially traced, the vertices become "Aichiken" and become the accepted vertices indicated by arrow B. Therefore, "Aichiken (Aichi Prefecture)" is one recognition target word. Then, through the reception vertex indicated by the arrow B, “Kariyashi” is formed and the reception vertex indicated by the arrow C is obtained. Therefore, "Aichi Ken Kariyashi (Kariya City, Aichi Prefecture)" is also one recognition target word. Further, there is a vertex of “child” via the receiving vertex indicated by the arrow C. FIG.
Although it is not shown in the figure, since it can be traced as "showacho" and has an acceptance vertex, "aichikenkariya shishowacho (Showa-cho, Kariya city, Aichi prefecture)" is one recognition target word.

In this case, for example, one recognition target word such as “Aichiken Kariyashi Showacho” (Kawatani City, Aichi Prefecture) is composed of “Aichiken (Aichi Prefecture)” and “Kariyashi (Kariya City). )) And Showacho (Showa-cho) are hierarchically connected. Therefore, it can be seen that there are three hierarchies by the existence of the reception vertices indicated by arrows B and C in FIG. In other words, the accepted vertex indicates that a word is composed of the syllable data that has been traced to it. Conversely, if there is more syllable data downstream from the accepted vertex, the upstream side of the accepted vertex is in the upper hierarchy. And the downstream side is the lower hierarchy. For example, consider the receiving vertex indicated by arrow B in FIG.
"Aichiken (Aichi Prefecture)" on the upstream side is the upper hierarchy,
"Kalyashi (Kariya City) ..." on the downstream side is the lower hierarchy. That is, in this case, the word indicating the prefecture is the upper hierarchy, and the word indicating the city level or lower is the lower hierarchy. In addition, considering the reception vertex indicated by arrow C in FIG. 3, “Aichiken Kariyashi (Kariya City, Aichi Prefecture)” on the upstream side is the upper hierarchy, and is not shown in FIG. "Wachi (Showa-cho)" is the lower hierarchy.

In the above description, the dictionary data stored in the dictionary unit 31b has been described with reference to the specific example of Showa-cho, Kariya city, Aichi prefecture.
Data is set for other place names with the city level as the second level and the town level as the third level. In addition,
"Basically" means "town" at the next city level in the prefecture
And "villages" come in some places.

The description of the dictionary section 31b is completed here. As described above, in the voice recognition unit 31, the matching unit 31a compares the voice data acquired from the voice input unit 34 with the plurality of comparison target pattern candidates stored in the dictionary unit 31b to determine the degree of matching. Is output to the storage unit 32a of the interaction control unit 32. The storage unit 32a stores the upper comparison target pattern. The input unit 32b inputs an external situation from the control circuit 10, and the inputted external situation is also stored in the storage unit 32a. The external situation in this case is data on the current position based on the position detected by the position detector 4 described above. It should be noted that the upper comparison target pattern stored in the storage unit 32a and the current location data as the input external situation are cleared when a predetermined confirmation instruction is given.

In the post-processing section 32c, for example, when the above-mentioned predetermined confirmation instruction is given, the data is sent to the control circuit 10 to execute "predetermination processing" for instructing the control circuit 10 to perform a predetermined processing. A process for instructing the voice synthesizing unit 33 to transmit the voice data and generate a sound is executed.
The data to be sent to the control circuit 10 in this case includes
All of the upper comparison target patterns as the final recognition result may be used, or only the uppermost pattern may be used.

On the other hand, in the configuration shown in FIG. 2B, the voice recognition section 31 is composed of a collation section 131a and a dictionary section 131b, and the dialogue control section 32 is composed of a storage section 132a and an input section 1a.
32b, a post-processing unit 132c, and a dictionary control unit 132d. In the configuration shown in FIG. 2A, if there is data to be referred to in the storage unit 32a of the dialog control unit 32,
The matching unit 31a of the voice recognition unit 31 creates an upper layer list that can be omitted by referring to the reference data stored in the storage unit 32a of the dialog control unit 32, and performs the matching process based on the list. In the configuration of FIG. 2B, the dialogue control unit 3
The second dictionary control unit 132d controls the dictionary data stored in the dictionary 31b. Specifically, the storage unit 132
A list of upper layers that can be omitted is created with reference to the reference data stored in a, and the upper layer part that can be omitted is temporarily omitted from the dictionary data stored in the dictionary unit 31b. The collation unit 31a executes the collation processing using the dictionary data thus controlled.

The operation of the post-processing unit 132c is the same as that in the case of FIG. Next, the operation of the car navigation system 2 of the first embodiment will be described. In addition, since a portion related to the voice recognition device 30 is a feature, a general operation as a car navigation system will be briefly described, and then, an operation of a portion related to the voice recognition device 30 will be described in detail.

After the power of the car navigation system 2 is turned on, the driver can change the guide route from the menu displayed on the display device 14 by using the remote controller 15a (the same operation can be performed with the operation switches 8; the same applies to the following description). When the route information display process is selected to be displayed on the display device 14, or when a desired menu is voice-inputted via the microphone 35 via the voice recognition device 30, the dialogue control unit 32 sends a message to the control circuit 10. When the same instruction is given via the remote controller 15a, the following processing is performed.

That is, when the driver inputs a destination by voice or an operation of a remote controller or the like based on a map on the display device 14, the current position of the vehicle is obtained based on satellite data obtained from the GPS receiver 22. A process is performed between the ground and the current location to calculate the cost by the Dijkstra method and determine the shortest route from the current location to the destination as a guidance route. And the display device 1
4. A guidance route is displayed on the road map above to guide the driver to an appropriate route. The calculation process and the guidance process for obtaining such a guide route are generally well-known processes, and a description thereof will be omitted.

Next, the operation of the voice recognition device 30 will be described by taking as an example a case where the destination for the route guidance is input by voice. FIG. 4 is a flowchart showing processing in the speech recognition unit 31 and the dialog control unit 32 in that case. Note that this flowchart shows processing on the assumption that the voice recognition unit 31 and the dialog control unit 32 have the configuration shown in FIG.

In the first step S10, it is determined whether or not there is a voice input. As mentioned above, PT
Since only the voice data input through the microphone 35 while the T switch 36 is pressed is output from the voice input unit 34 to the voice recognition unit 31, it is determined whether there is a voice input from the voice input unit 34. Judge. If there is a voice input (S10: YES), the storage unit 32a in the dialog control unit 32
With reference to FIG. 2A (S20), it is determined whether there is reference data (S30). This reference data has been stored in the process of S80. The processing of S80 will be described later. If there is reference data in the storage unit 32a (S30: YES), the following two processes are executed based on the reference data (S40). That is, creation of an upper layer list that can be omitted and determination of a vertex to start collation. Here, a case will be described in which “Aichikenkariya Shishouichi Ichigo (Matsuei-cho, Kariya-shi, Aichi)” which is a specific example of the dictionary data shown in FIG. 3 is stored as reference data. Regarding this reference data, considering the above-mentioned reception vertices B and C (see FIG. 3), the upper hierarchy “Aichiken (Aichi Prefecture)” and the upper hierarchy “Aichiken Kariyashi (Kariya City, Aichi Prefecture)” There are two possibilities. Therefore, in this S40, "Aichiken" and "Aichiken Kariyashi" are created as the above-mentioned optional upper hierarchy list. Then, the matching start vertex is determined as follows. That is, in the case of the dictionary data shown in FIG. 3, as described above, the vertex indicated by the arrow A in FIG. 3 becomes the “root”, and the syllables are allocated to each side according to preorder traversal from there. The word is completed by tracing. Therefore, in this state, the collation start vertex is only the vertex indicated by the arrow A, but two omissible upper-layer lists are created as described above, and the collation start vertex corresponding thereto is additionally determined. In other words, the reception vertex indicated by arrow B, which is the reception vertex of the upper hierarchy list “Aichiken”, and the reception vertex indicated by arrow C, which is the reception vertex of the upper hierarchy list “Aichiken Kariyashi”, are determined as new verification start vertices. You do it.

After the processing in S40, the flow shifts to S50, where the voice recognition processing is executed. If a negative determination is made in S30, that is, if there is no reference data, the process proceeds to S50 without executing the process of S40. Here, the speech recognition processing in S50 will be described. Basically, matching is performed on the acquired voice data using the dictionary data stored in the dictionary unit 31b. At this time, since the matching start vertex is additionally determined in S40, the matching is performed. Words starting from the additionally determined collation start vertex are also temporarily determined to be words in the dictionary data and collation is performed. If the upper comparison target pattern is determined based on the result of the comparison with the word that starts from the additionally determined collation start vertex, the corresponding one of the upper hierarchy list stored in the storage unit 32a is compared with the upper comparison pattern. After being added to the target pattern, a higher-level comparison target pattern as a final recognition result is determined. Thereby, the recognition result is determined. For example, when the accepted vertex indicated by arrow C in FIG. 3 is the matching start vertex, and the word “showacho (Showa-cho)” starting from the matching start vertex is determined as the upper comparison target pattern, The upper level list is "Aichi Ken Kariyashi"
Therefore, the final recognition result is "Aichi Kenkari Yasushi Shocho (Showa-cho, Kariya City, Aichi Prefecture)".

Returning to the description of the flowchart of FIG. 4, in S60, it is determined whether or not the recognition result of the voice recognition processing in S50 is a voice input of "Yes". Then, if the voice input is not “yes” (S6
0: NO), and in S70, it is determined whether or not the recognition result belongs to a predetermined category. Here, since it is premised on the process of setting a destination for route guidance, the predetermined category is a category relating to a place name.

In the case of this predetermined category (S70: YE
S), the process proceeds to S80, and the recognition result is stored as reference data in the storage unit 32a (see FIG. 2A). Then, in S90, a voice response process is executed. This is a process of outputting the recognition result as voice via the voice synthesis unit 33 and the speaker 37.

On the other hand, if the category is not the predetermined category (S70:
NO), and the process proceeds to S100 to execute other processing.
After the processing of S90 or S100, return to S10,
Repeat the process. If a positive determination is made in S60, that is, if the recognition result is a voice input of “Yes”, the process proceeds to S60.
The process proceeds to step 110 to determine the recognition result. And the following S1
At 20, a predetermined post-determination process is executed. The post-confirmation processing in this case means that data relating to “destination for route guidance” as a recognition result is sent to the control circuit 10 (see FIG. 1).
To output to

After such post-determination processing is completed, S
The process proceeds to S130, where the reference data stored in the storage unit 32a in the process of S80 is deleted. Then, the process returns to S10. The above is the description of the operation in the case of voice input of the destination for route guidance as an example, but in order to more clearly understand the features and effects related to voice recognition of the present invention,
As mentioned in the processing description of S50, the description will be continued with a specific example in which "Showa-cho, Kariya-shi, Aichi" is specified as the destination.

It is assumed that the user has voice-inputted “Showa-cho, Kariya-shi, Aichi” through the microphone 35. Speech recognition accuracy is 1
If it is not 00%, it may be erroneously recognized.
For example, if “Shoei town, Kariya city, Aichi prefecture” is mistakenly recognized, the voice recognition device 30 outputs the voice via the speaker 37.

As a result, the user knows that the user is erroneously recognized. Therefore, the user must again input "Showa-cho, Kariya-shi, Aichi" to the voice recognition device 30 side. However, when the user inputs voice again, instead of inputting voice as "Showa-cho, Kariya-shi, Aichi Prefecture", only voice input of "Showa-cho" is sufficient. In other words, the first voice input, "Showa-cho, Kariya-shi, Aichi" is a hierarchical connection of multiple words indicating different levels of place names, "Aichi-ken", "Kariya-shi", and "Showa-cho". Therefore, the words or words constituting the upper hierarchy, "Aichi Prefecture" or "Kariya City, Aichi Prefecture" are omitted in the recognition process (the upper hierarchy list that can be omitted described in S40 of FIG. 4). ). Therefore, even if the user only inputs a voice for "Showa-cho" for the second time, it is possible to perform comparison with the input voice on the assumption that the input is "Kariya-shi, Aichi". of course,
It is also possible to respond to the situation where only Aichi Prefecture is omitted and the voice is input again as "Showa-cho in Kariya city".

As described above, when the speech recognition device 30 misrecognizes the character, correcting only the misrecognized part (the town name "Showa-cho" in the above example) is performed in a conversation in daily life. It is very natural in light of his habits. Forcing the user to pay special attention only when using the voice recognition device 30 is not preferable in terms of usability. Therefore, since it is possible to cope with a very natural behavior in the daily conversation habit of correcting only the part of erroneous recognition as in the present speech recognition device 30, even if it is more natural to omit the upper layer, The user's burden of having to input voice can be reduced, and usability can be further improved.

As shown in the flowchart of FIG. 4, when the post-determination processing in S120 is completed, S1
At 30, the reference data in the storage unit 32a is deleted. That is, assuming that the reference data is erroneously recognized, S
Since it is stored for the creation of an optional upper layer list in 40, once the post-determination processing in S120 is completed, it is no longer necessary, and conversely, Since it is irrelevant data, it is deleted here.

By the way, the processing contents shown in the flowchart of FIG. 4 are based on the configuration shown in FIG. 2A, but the case where the configuration shown in FIG. deep. Also in this case, the basic processing is the same, but the processing contents of S40 and S50 are slightly different. In other words, on the premise of the configuration shown in FIG. 2A, an upper layer list that can be omitted and a vertex to start collation are determined in S40, and in the voice recognition process in S50, an additional determination is made in S40. Words starting from the matching start vertex are also temporarily regarded as words in the dictionary data, and the matching is performed.

On the other hand, assuming the configuration shown in FIG. 2B, the following processing is executed as processing corresponding to S40. That is, although the creation of the upper-level list that can be omitted is the same, the dictionary control unit 132d
Is based on the optional upper hierarchy list,
A second comparison target pattern candidate is obtained by omitting the words constituting the upper layer list from the corresponding comparison target pattern candidate stored in 1b. Speaking of the specific examples described above, "showacho (Showa-cho)" and "Kariyashi Showacho (Showa-cho in Kariya city)" correspond to this. Therefore, as a process corresponding to S50 in FIG. 4, the dictionary control unit 132d uses the comparison target pattern candidate stored in the dictionary unit 31b in a state where the second comparison target pattern candidate is newly established.
Is performed. In this case,
The processing corresponding to S130 in FIG.
Is deleted, and the dictionary comparison unit 132d deletes the second comparison target pattern candidate newly installed in the dictionary unit 31b, and returns to the original state. [Another Embodiment] One embodiment has been described above with reference to FIGS. 1 to 4, but some other embodiments will be described.

(1) In the above embodiment, when a speech input belonging to the same predetermined category as the previous recognition result is made within a predetermined period after notifying the recognition result as an object to be omitted in the recognition processing, Words that constitute the upper layer of the previous recognition result were assumed. However, the present invention is not limited to this, and the following embodiments are also conceivable. That is, this is a case based on the external situation (see FIG. 2) input from the control circuit 10.
More specifically, the information is current position information or the like input from the navigation device body in the car navigation system 2. That is, when a vehicle equipped with the car navigation system 2 is traveling in, for example, Aichi Prefecture, and "Awa Prefecture Kariya City Showa Town" in the same Aichi Prefecture is set as the destination, "Showa Town Kariya City Aichi Prefecture" It is more natural to omit the word "Aichi" and input the word "Showa-cho Kariya" instead of inputting the voice. In other words, there is a situation in which the user thinks that designation from the city level, which is the lower hierarchy, is sufficient even if the user does not specify from Aichi Prefecture.

In such a case, the dialog control unit 32 may input the current location as the external situation from the control circuit 10 and set, for example, the upper hierarchy "Aichi prefecture" as an omission target in the recognition processing. As described above, in a predetermined area in which the user is currently located, a place in the upper hierarchy for specifying the area is not dared to be specified, but is specified from the lower hierarchy. By being able to cope with the behavior, even if it is more natural to omit the upper layer, it is possible to reduce the user's burden of having to input voice from the upper layer, and to further improve the usability.

(2) In the above embodiment, the speaker 37 is used as the "notifying means", and the recognition result is notified by voice output. This eliminates the need for the driver to move his / her viewpoint, thereby contributing to safer driving. That is, since the car navigation system 2 is used for in-vehicle devices, there is an advantage in such audio output. Of course, the recognition result may be notified by displaying a character or a symbol on the screen, or may be notified by outputting a voice and may be notified by displaying the character or the symbol on the screen. . Then, when displaying the recognition result on the screen, the display device 1 for displaying the map information of the car navigation system 2
4 (see FIG. 1).

In the process of S40 in FIG. 4, it is determined whether or not the recognition result belongs to a predetermined category. The predetermined category is a category related to a place name on the assumption that a destination is set. Was explained. However, the gist of the present invention is not limited to such place names and the like, but in abstract terms, information that needs to be formally output after outputting a recognition result and obtaining confirmation from a user. Category. More specifically, in the case of the car navigation system 2 described above, this means an instruction of predetermined navigation processing related information that needs to be specified in performing the navigation processing. A representative one of the “predetermined navigation processing related information” is a destination, but other instructions that need to be designated in performing the navigation processing, such as selecting conditions for a route search, are also included.

As described above, the present invention is not limited to such an embodiment, and can be implemented in various forms without departing from the gist of the present invention. For example, in the above-described embodiment, the example in which the voice recognition device 30 is applied to the car navigation system 2 has been described, but the application destination is not limited to the above-described car navigation system 2. For example, when using a voice recognition device for an air conditioning system, adjustment of the set temperature, air conditioning mode (cooling /
Heating / drying) or the wind direction mode may be selected by voice input. For example, as for the set temperature, there are instructions related to the set temperature, such as "set the set temperature to 25 degrees" or "decrease the set temperature by 5 degrees", but there are a plurality of instructions. According to the present invention, if the user inputs "set the set temperature to 25 degrees" but mistakenly recognizes "set the set temperature to 22 degrees", the input of "25 degrees" will be "25."
It is only necessary to input "to the degree", and the usability of the user is also improved. The same applies to the air conditioning mode and the wind direction mode.

The car navigation system 2 and the air conditioning system described above are not limited to the case where they are used as in-vehicle equipment, but may be, for example, a portable navigation device or an indoor air conditioner. However, as described above, when used for in-vehicle equipment, the user may be a driver. In that case, driving itself is the most important. For other in-vehicle equipment, drive as much as possible. Preferably, there is no hindrance. Therefore, in the case of a voice recognition device on the premise of the car navigation system 2 or the air conditioning system as on-vehicle equipment, there is a further advantage. Of course, from such a viewpoint, the present invention can be similarly applied to in-vehicle devices other than the navigation system and the air conditioning system. For example, a car audio device is effective. Also, considering a configuration in which opening and closing of the power window and adjustment of the mirror angle are instructed by voice, it is effective even in such a situation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a car navigation system as an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a speech recognition unit and a dialogue control unit in the speech recognition device.

FIG. 3 is an explanatory diagram showing dictionary data stored in a dictionary unit in a voice recognition unit.

FIG. 4 is a flowchart illustrating processing related to voice recognition and dialog control in the voice recognition device.

[Explanation of symbols]

2 ... Car navigation system 4 ... Position detector 6 ... Map data input device 8 ... Operation switch group 10 ... Control circuit 12 ... External memory 14 ... Display device 15 ... Remote control sensor 15a ... Remote control 16 ... Geomagnetic sensor 18 ... Gyroscope 20 ... Distance sensor 22 GPS receiver 30 Voice recognition device 31 Voice recognition unit 31a Matching unit 31b Dictionary unit 32 Dialogue control unit 32a Storage unit 32b Input unit 32c Post-processing unit 33 Voice synthesis unit 34 Voice input unit 35 Microphone 36 PTT switch 37 Speaker 132a Storage unit 132b Input unit 132c Post-processing unit 132d Dictionary control unit

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FI G09B 29/10 G09B 29/10 A

Claims (9)

[Claims] 1. A voice input unit for inputting voice, and a voice input via the voice input unit is compared with a plurality of comparison target pattern candidates stored in advance in a dictionary unit to determine a matching degree. A recognizing means for recognizing a higher result, a notifying means for notifying the recognition result by the recognizing means, and a predetermined confirmation instruction after the recognizing result is notified by the notifying means, the recognizing result is displayed. And a post-confirmation processing means for executing predetermined post-confirmation processing as being determined, wherein at least one of the plurality of comparison target pattern candidates stored in the dictionary means is a plurality of Are set as hierarchically connected words, and a word or a group of words constituting an upper layer of the comparison target pattern candidate obtained by hierarchically connecting the plurality of words is recognized. When set as an omission target, the recognizing means temporarily omit the set upper layer constituent word or word group as the comparison target pattern candidate, A speech recognition device configured to perform a comparison on the input speech. 2. The speech recognition device according to claim 1, wherein a speech input belonging to the same predetermined category as a previous recognition result is made within a predetermined period after notifying the recognition result. A speech recognition device, wherein a word or a group of words constituting an upper hierarchy of a recognition result is set as an omission target in the recognition processing. 3. The speech recognition apparatus according to claim 2, wherein said recognition means temporarily omits a word constituting a higher hierarchical level from a previous recognition result and temporarily considers the word as a comparison target pattern candidate. The speech recognition device according to claim 1, wherein the predetermined period during which the comparison with the current input speech is permitted is performed until the predetermined confirmation instruction is issued after the notification of the recognition result. 4. The speech recognition apparatus according to claim 1, wherein a word or a group of words constituting an upper layer set as an object to be omitted in the recognition processing uses a speech recognition result by the speech recognition apparatus. A voice recognition device which is input from a device other than the voice recognition device in the system. 5. The speech recognition device according to claim 1, wherein upper-level constituent words or word groups set as omission objects in the recognition processing are temporarily stored as reference data. Reference data storage means, wherein the recognition means performs a recognition process, from a plurality of comparison target pattern candidates stored in the dictionary means, a higher hierarchy of reference data stored in the reference data storage means. Is also regarded as a candidate for the comparison target pattern, and then a comparison with the current input speech is executed. 6. The speech recognition apparatus according to claim 1, wherein, if the previous recognition result is obtained by connecting a plurality of words hierarchically, the recognition result is used as reference data. A reference data storage means for temporarily storing, and a higher layer of the reference data is constructed from the corresponding comparison target pattern candidates stored in the dictionary means based on the reference data stored in the reference data storage means. Dictionary control means for newly establishing a second comparison target pattern candidate in which words to be omitted are provided. The dictionary control means stores the second comparison target pattern candidate in the newly established dictionary means. A speech recognition device, wherein the recognition unit is configured to perform a comparison with a current input speech using a comparison target pattern candidate. 7. The voice recognition device according to claim 1, wherein the notification unit is a unit that outputs a voice to notify the user, and the notification of the recognition result is a content of the recognition result. Is output from the voice output unit as a voice. 8. A voice recognition device according to claim 1, further comprising: a navigation device, wherein the voice input unit of the voice recognition device performs at least navigation processing by the navigation device. The user uses a voice to input an instruction of predetermined navigation processing-related data that needs to be specified, and the post-confirmation processing means sends the recognition result by the recognition means to the navigation device. A navigation system configured to output. 9. A voice recognition device according to claim 4, further comprising a navigation device, wherein the voice input means of the voice recognition device needs to be specified at least when the navigation device performs a navigation process. It is used for a user to input an instruction of certain place name related data by voice, and the post-confirmation processing means is configured to output a recognition result by the recognition means to the navigation device. A navigation system, wherein the navigation device includes: a current position detection unit that detects a current position; and a current position detected by the current position detection unit, when the current position detected by the current position detection unit can be represented as the comparison target pattern candidate in which a plurality of words are hierarchically connected. The speech recognition apparatus as a word or a group of words constituting an upper layer thereof, Navigation system characterized in that it comprises, and optional target data output means for outputting.