JP3500948B2 - Voice recognition device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device suitable for being incorporated in a navigation system, for example, and relates to a voice recognition device that enables voice input when inputting a destination or the like to the navigation system.

[0002]

2. Description of the Related Art A voice recognition device of this type inputs a voice uttered by a user, compares the input voice with a plurality of comparison target pattern candidates stored in a recognition dictionary, and compares them. It is configured to calculate the degree of coincidence between the user generated voice and the comparison target pattern candidate, that is, the degree of similarity. The comparison target pattern candidate having the highest similarity among the calculated similarities is output as the recognition result.

When the voice recognition device having such a configuration is incorporated in a navigation system, when a destination or the like is input to the navigation system, it is possible to input by voice. In the case of this configuration, the navigation system can be operated by voice, which is a convenient device that is considerably easy to use even for a driving user.

[0004]

However, in the above-described conventional voice recognition device, when a voice recognition is performed for a destination or a name having similar pronunciation, the recognition result is sometimes incorrect. When such misrecognition occurs, a process unintended by the user may be executed, for example, a destination unintended by the user may be set.In such a case, the user is uncomfortable. It was

Therefore, an object of the present invention is to provide a voice recognition device capable of improving the recognition rate of voice recognition and preventing the user from feeling uncomfortable as much as possible.

[0006]

According to the invention of claim 1,
Input voice, prefecture name, city name, position, type
And multiple comparison targets with rank data added
The similarity is calculated by comparing with the pattern candidates, and the similarity is calculated.
Recognition that is configured so that the recognition result is high
Is it data indicating a certain point or a certain area on the device ?
An information input means for inputting Ranaru priority recognition information, said Yu
It is added to the previous recognition information and the plurality of comparison target pattern candidates.
Comparison result of the positional information obtained, the plurality of comparison objects
Using the type and rank added to pattern candidates
Adjustment rule for adjusting the similarity and the priority recognition information
And the prefecture name, city name, location, type and
Patterns to be compared with the linked data
Apply the candidate to the adjustment rule to adjust the similarity
And a similarity adjusting means. In this configuration
According to the comparison, the comparison should be recognized first by the user.
The target pattern candidates are easily recognized, and the recognition rate is further improved.

In this case, as in the invention of claim 2, it is configured to input the priority recognition information by an input operation by the user, and, like the invention of claim 3, a memory for storing the priority recognition information. It is preferable that the device is provided with the means. With this configuration, the recognition rate can be improved so as to match the user's wishes.

According to the fourth aspect of the invention, a position detecting means for detecting the current position is provided, and the detected current position is inputted as the priority recognition information. According to this configuration, the recognition rate is automatically improved, so that the usability for the user can be further improved.

[0009]

According to the fifth aspect of the present invention, since the adjustment rule is changed or increased or decreased by the learning process, the recognition rate is automatically improved. In this case, as in the invention of claim 6 is preferably configured to increased or decreased to change the adjustment rule based on the recognition history. With this configuration, the recognition rate is further improved.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to a car navigation system will be described below with reference to the drawings. First, FIG. 2 is a block diagram schematically showing the overall configuration of the car navigation system 1. As shown in FIG. 2, the car navigation system 1
It is configured to include a voice recognition device 2 and a navigation device 3. The voice recognition device 2 includes a microphone 4 and a PT.
The T (Push-To-Talk) switch 5 and the speaker 6 are connected.

Further, a specific configuration of the navigation device 3 is shown in FIG. In FIG. 3, the control circuit 7 of the navigation device 3 is configured to include a microcomputer and has a function of controlling the overall operation of the navigation device 3. A position detector 8, a map data input device 9, an operation switch group 10, an external memory 11, a display device 12, and a remote control sensor 13 are connected to the control circuit 7. Further, the control circuit 7 is connected to (the control unit 14 (see FIG. 4) of) the voice input device 2 described above.

The position detector 8 is composed of a combination of a geomagnetic sensor, a gyroscope, a distance sensor, a GPS receiver (none of which is shown), or a part thereof. The position detector 8 is configured to detect the current position of a vehicle equipped with the car navigation system 1 of this embodiment and output a current position detection signal. In this case, the position detector 8 constitutes the position detecting means and the information inputting means. Also, the map data input device 9
Is a device for inputting map data, map matching data, and the like. The data such as the map data is recorded on a recording medium such as a CD-ROM.

Further, the display device 12 is composed of, for example, a liquid crystal display or the like, and can display in color so that a map or the like can be clearly displayed. Operation switch group 10
Is composed of a touch switch (touch panel) provided on the upper surface of the screen of the display device 12, a mechanical push switch provided on the peripheral portion of the screen, and the like. The remote controller sensor 13 is a receiver that receives a transmission signal transmitted from the remote controller 13a operated by the user. Then, the control circuit 7 automatically selects and sets the optimum route from the current position to the destination when the user sets the destination by operating the operation switch group 10 or the remote controller 13a. It has a function of executing a map matching process for locating a position on a map.

When setting the destination and the like, the user can use the voice recognition device 2 to input a voice instead of operating the operation switch group 10 and the remote controller 13a. . Hereinafter, the voice recognition device 2 will be described with reference to FIG.

As shown in FIG. 4, the voice recognition device 2 comprises a control unit 14, a voice input unit 15, a voice recognition unit 16 and a voice synthesis unit 17. Here, the control unit 14
It has a function of controlling the overall operation of the voice recognition device 2. The control unit 14 is connected to the control circuit 7 of the navigation device 3, and is configured to exchange data with the control circuit 7.

The voice input unit 15 is also configured to input a voice uttered by a user through the microphone 4 and output voice data (eg, digital data) to the voice recognition unit 16. In this case, the voice input unit 15 is configured to output the voice data to the voice recognition unit 16 only while the user depresses the PTT switch 8. That is, the voice recognition processing of the voice uttered by the user is executed only while the PTT switch 8 is operated.

Then, the voice recognition unit 16 performs voice recognition processing on the voice data (input voice) given from the voice input unit 15 according to an instruction from the control unit 14, and outputs the voice recognition result to the control unit 14. It is configured to output. The voice recognition unit 16 is specifically composed of a collation unit 18 and a dictionary unit 19, as shown in FIG. The dictionary unit 19 stores dictionary data composed of a recognition target vocabulary and a tree structure (known data structure) of the recognition target vocabulary. Here, in the recognition target vocabulary, that is, the data of the comparison target pattern candidate, as shown in FIG. 5, data of prefectures, municipalities, recognition vocabularies, positions, types and ranks (so-called attribute data) is included. Has been added. FIG. 5 shows attribute data for two vocabularies “Tokyo Disneyland” and “Kariya Post Office” as recognition target vocabularies. The attribute data is added to other recognition target words in the dictionary data in substantially the same manner.

The collating unit 18 is also configured to collate (recognize) the voice data given from the voice input unit 15 by using the dictionary data stored in the dictionary unit 19. . In this case, first, the voice data and the plurality of comparison target pattern candidates in the dictionary data are compared to calculate the similarity (that is, a value obtained by calculating the degree of coincidence between the two). The process of calculating the degree of similarity is executed using a control program (algorithm) for collation process that is already known.

Next, for example, the current position (the current position detected by the position detector 8) which is the priority recognition information,
Adjustment is performed to increase or decrease the value of the similarity based on the adjustment rule stored in the storage unit 20 (see FIG. 1) provided in the control unit 14. Then, the matching unit 18 is configured to output the comparison target pattern candidate (recognition target vocabulary) having the highest adjusted similarity to the control unit 14 as a recognition result. In this case, the collating unit 18 constitutes a similarity adjusting unit. The specific content of the process of adjusting the degree of similarity will be described later.

The adjustment rules stored in the storage unit 20 of the control unit 14 include standard rules preset by the manufacturer and rules entered by the user. Will be described later. And
When the user inputs a rule, the user operates the operation switch group 10 and the remote controller 13a of the navigation device 3 to input a desired rule. When the user inputs a rule, the rule may be input by voice.

On the other hand, when the voice synthesizing unit 17 receives from the control unit 14 data (for example, text data composed of kana characters) representing a voice to be uttered, it is configured to synthesize a voice from this voice data. ing. Then, the voice synthesizing unit 17 outputs the synthesized voice to the speaker 6
It is configured to be output from and uttered.

Next, the operation of the above configuration, specifically, the operation when the user sets the destination by voice,
The description will be made with reference to FIGS. 6 and 7. The flowchart of FIG. 7 shows the contents of the control portion that executes the voice recognition processing of the control program that operates the voice recognition device 2.

First, while pressing the PTT switch 5, the user speaks into the microphone 4 the name of the destination to be set (for example, "Tokyo Disneyland"). Then, the voice input unit 15 of the voice recognition device 2 inputs the voice uttered by the user through the microphone 4 and outputs the voice data to the voice recognition unit 16. As a result, the voice input is acquired in the voice recognition device 2 (step S1 in FIG. 7).

Next, the voice recognition unit 16 of the voice recognition device 2
Acquires, for example, information indicating the current position as the priority recognition information (step S2). In this case, the voice recognition unit 16
Is configured to receive information (current position information) indicating the current position detected by the position detector 8 of the navigation device 3 via the control circuit 7 and the control unit 14 of the voice recognition device 2. The information indicating the current position acquired here is data indicating a predetermined area to which the current position belongs, such as “Kariya City, Aichi Prefecture”. still,
As the information indicating the current position, for example, it is preferable that the coordinates of the current position (longitude and latitude) are also acquired together.

After that, the collation unit 18 of the voice recognition unit 16
Data matching (recognition) is performed on the voice data provided from the voice input unit 15 using the dictionary data stored in the dictionary unit 19 (step S3). Here, the degree of similarity (that is, the degree of coincidence) is calculated by comparing the voice data and the plurality of comparison target pattern candidates in the dictionary data. In this case, for the voice uttered by the user, for example, "Tokyo Disneyland", the following 3
Assume that one recognition candidate is extracted. The first candidate is "Tokyo XX Post Office", which has a rank of B and a similarity of 1
It is assumed that the score is 00. The second candidate is “Tokyo Disneyland”, which is ranked A and has a similarity of 95 points. The third candidate is "Tokyo XX Land", which is ranked C and has a similarity of 90 points.

Here, in the conventional speech recognition apparatus, the one having the highest degree of similarity among the above three candidates, that is,
“Tokyo Post Office” is output as a recognition result. On the other hand, in the voice recognition device 2 of the present embodiment, the matching unit 18 uses the data of the current position and the adjustment rule stored in the storage unit 20 of the control unit 14 to calculate the similarity. Is configured to increase or decrease the value of (step S4).

Specifically, as shown in the table of FIG. 6, the value of the degree of similarity is adjusted in consideration of, for example, three conditions. Here, the first condition of the above three conditions is whether the recognition candidate word exists outside the adjacent prefecture (or is it in the same municipality), or the recognition candidate word exists and the current position. It is determined whether or not the distance between and is ∘km or more (or is within ∘km). In this case, ∘km is a preset setting value. The second condition is to determine what kind (genre) of the recognition candidate word is, for example, an amusement park, a post office, ... The third condition is to judge whether the rank of the recognition candidate word is A, B, or C. Here, rank A represents a well-known facility, rank B represents a moderately known facility, and rank C represents a minor facility (a facility known only to the locals who have the facility). ) Is represented.

Then, the matching unit 18 determines the similarity value of the extracted recognition candidates while taking the above three conditions into consideration.
The adjustment is configured as shown in the item (treatment) at the right end of the table in FIG. Specifically, if it is rank A in an amusement park outside the adjacent prefecture, 10 is added to the similarity (see No. 1 in FIG. 6). If the rank is C in an amusement park outside the adjacent prefecture, 10 is subtracted from the similarity (see No. 2 in FIG. 6). If the rank is B at a post office outside the adjacent prefecture, the similarity is subtracted by 5 (see No. 3 in FIG. 6). If the rank is B at the post office in the same municipality, 5 is added to the similarity (see No. 4 in FIG. 6). The adjustment rule for adjusting the degree of similarity as shown in the table of FIG. 6 is configured to be stored in the storage unit 20 of the control unit 14 in advance.

Here, the reason why the above adjustment rule is set will be briefly described. First, a well-known (famous) facility (rank A) such as "Tokyo Disneyland" is likely to be set as a destination from anywhere in the country, so it is necessary to make it easy to be recognized regardless of the current location. Therefore, the degree of similarity is adjusted to be high. On the other hand, local facilities closely related to daily life such as "XX public health center" and "XX post office" (rank B)
Since the use of local residents is overwhelming, when the current position is close (for example, in the same municipality), the degree of similarity is adjusted to be high to facilitate recognition. On the other hand, when the local facility is far away from the current position (when it is outside the adjacent prefecture), the similarity is adjusted to be low so that it is difficult to be recognized.

In this case, since the user hardly knows information about the area that has never been visited (name of facility, etc.),
It is preferable that the facility in such a region is not recognized, and the facility in the region becomes difficult to be recognized by the adjustment processing, which is preferable in accordance with the usability of the user. On the other hand, since the user knows well-known facilities and local information (names of facilities, etc.) in detail, it is preferable that such well-known facilities and local facilities are recognized. The facility is easy to be recognized, which is preferable according to the user's feeling of use.

Next, when the similarity of each recognition candidate word is adjusted as described above, the similarity of "Tokyo Disneyland" among the above three recognition candidate words becomes the highest. As a result, "Tokyo Disneyland" is output as the recognition result (step S5).

Now, in the above-mentioned adjustment of the degree of similarity,
The post office's similarity was reduced, so that the post office was adjusted to be difficult to recognize. On the other hand, when the current position is Kariya city, Aichi prefecture, when the user says "Kariya post office", the similarity of "Kariya post office" as a recognition candidate is adjusted to be added. . Therefore, "Kariya Post Office" is easily recognized as a recognition result, and the recognition rate is improved, so that the user's feeling of use is improved.

According to this embodiment having such a configuration, for example, the current position is input as the priority recognition information, and the recognition candidate word (comparison target pattern) is calculated based on this current position and the adjustment rule stored in the storage unit 20. Since the similarity of the (candidate) is adjusted, the similarity of the recognition candidate word which should be preferentially recognized by the user is increased, and the recognition candidate word is easily recognized. As a result, the recognition rate of voice recognition is improved, and it is possible to minimize discomfort to the user.

Further, in the above embodiment, the position detector 8 for detecting the current position is provided, and the detected current position is inputted as the priority recognition information. With this configuration, it is almost impossible to recognize a facility in a remote area that has never been visited by a user, and the facility is not well known, and it is automatically recognized even when the car moves. Since the rate is improved, the usability of the user can be further improved.

Further, in the above-mentioned embodiment, the data of the prefecture, the name of the municipality, the position, the type and the rank are added to the recognition candidate word (the pattern candidate for comparison), and the added data and the current position ( The similarity is adjusted based on the priority recognition information) and the adjustment rule. As a result, the degree of similarity can be adjusted more finely, and the recognition rate can be further improved.

In the above embodiment, the recognition candidate word (comparison target pattern candidate) may have a structure in which at least one of the prefecture name, the city name, the position, the type and the rank is added. Alternatively, it may be configured to add the data obtained by combining a plurality of appropriate data among them.

Further, in the above embodiment, the information of the current position detected by the position detector 8 is inputted as the priority recognition information, but the present invention is not limited to this.
For example, it may be configured such that information such as a residence of the user or a use place (use area) where the user mainly uses the vehicle is input by an input operation by the user. In this case, it is preferable to store the information (priority recognition information) such as the place of residence and the place of use of the user in the storage unit. As this storage means, EEPRO
A non-volatile memory such as M or a RAM with a backup function may be used. When inputting information such as a place of residence or a place of use, the operation switch group 10 of the navigation device 3, the remote controller 13a, or the like may be operated or input by voice. When the information such as the place of residence and the place of use is used as the priority recognition information, the recognition rate of the facility toward the home is improved even when the user travels far away.

Further, in the above embodiment, the adjustment rule stored in advance in the storage unit 20 is used to adjust the degree of similarity. However, the present invention is not limited to this, and the adjustment rule desired by the user may be used. Is stored (registered) in the storage unit 20.
Then, the adjustment rule may be used. In this case, the user operates the operation switch group 10 and the remote controller 13a of the navigation device 3 to store the desired adjustment rule in the storage unit 20 of the control unit 14 of the voice recognition device 2 via the control circuit 7. It may be configured so as to be stored (registered). Further, the adjustment rule desired by the user may be registered by voice. When the adjustment rule desired by the user can be freely registered in this way, the recognition rate is further improved.

In the case where the above embodiment is a voice recognition system in which, for example, "place name + facility name" is input by voice to specify a target object, the facility around the current position and the user are For facilities around the registered address and place, create an adjustment rule that allows you to specify the target object by simply inputting "facility name" by voice, and register this rule in the storage unit 20. Is preferred. With this configuration, the usability is further improved for the user.

Furthermore, it is also a preferable configuration to change or increase or decrease the adjustment rule by the learning process. As an example of the configuration of this learning process, a recognition history (for example, the number of times each facility name is actually recognized in the past) is stored, and the adjustment rules can be changed or the adjustment rules can be increased or decreased based on this recognition history. It is conceivable that the adjustment rule is changed so that the facility name that is recognized more frequently is more easily recognized. And
With this configuration, the recognition rate is automatically improved, and the recognition rate is further improved.

Although the voice recognition device 2 of the present invention is applied to the car navigation system 1 in the above embodiment, the present invention is not limited to this. For example, a vehicle air conditioner (so-called car air conditioner), a car audio device, It may be applied to an indoor air conditioner, a portable navigation device, or the like. Further, it is conceivable that the power window opening / closing operation command, the mirror reflection angle setting operation command, or the like may be configured to be performed by voice. Even if the voice recognition device of the present invention is applied to such a configuration good.

[Brief description of drawings]

FIG. 1 is a block diagram of a voice recognition unit showing an embodiment of the present invention.

FIG. 2 is a block diagram of a car navigation system.

FIG. 3 is a block diagram of a navigation device

FIG. 4 is a block diagram of a voice recognition device.

FIG. 5 is a diagram showing a data structure of a recognized vocabulary registered in dictionary data.

FIG. 6 is a diagram showing a table for explaining an example of adjustment rules.

FIG. 7: Flow chart

[Explanation of symbols]

1 is a car navigation system, 2 is a voice recognition device,
3 is a navigation device, 4 is a microphone, 5 is a PTT switch, 7 is a control circuit, 8 is a position detector (information input means,
(Position detection means), 9 is a map data input device, 10 is an operation switch group, 12 is a display device, 14 is a control unit, 15 is a voice input unit, 16 is a voice recognition unit, and 18 is a collation unit (similarity adjustment unit). , 19 is a dictionary unit, and 20 is a storage unit.

Claims (6)

(57) [Claims] [Claim 1] Input voice , prefecture name, city
Village name, location, type and rank data are added.
Compared plurality of the comparison target pattern candidates to calculate the similarity that, in the speech recognition device configured to be the result recognize what a high degree of similarity, the data indicating the predetermined point or a predetermined local Information input means for inputting priority recognition information consisting of, the priority recognition information and the plurality of comparison target pattern candidates.
The comparison result obtained by comparing the added position information, the plurality of ratios
The type and rank added to the comparison target pattern
Using the adjustment rule to adjust the similarity, the priority recognition information, the prefecture name, city name, place
Location, type, and rank data
Speech recognition apparatus characterized by a comparison pattern candidate example Bei the similarity adjusting means for adjusting the degree of similarity is applied to the adjustment rule. 2. The voice recognition device according to claim 1, wherein the information input unit is configured to input the priority recognition information in response to an input operation by a user. 3. The voice recognition device according to claim 2, further comprising a storage unit that stores the priority recognition information . 4. The voice according to claim 1, further comprising position detecting means for detecting a current position, wherein the information input means is configured to input the current position as the priority recognition information. Recognition device. 5. The adjustment rule is changed by a learning process.
The voice recognition device according to claim 1, wherein the voice recognition device is configured to increase or decrease . 6. The voice recognition device according to claim 5, wherein the adjustment rule is changed or increased / decreased based on a recognition history .