JP3291347B2 - 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 voice recognition apparatus for recognizing voice and controlling the apparatus based on the recognized result. The display state of the display section that is
The present invention relates to a speech recognition device that sequentially changes the vocabulary to be recognized according to the state change.

[0002]

2. Description of the Related Art Recent advances in computer technology have
Many attempts have been made to recognize voices uttered by humans and to control various devices based on the recognition results. Such attempts have been applied to telephones, car audio, or navigation systems.

For example, in a navigation system, various kinds of information are displayed on a display device, and control by voice recognition can be performed for each display scene. For this reason, it is general that the recognition target vocabulary for speech recognition is made different depending on the display state of the display device of the navigation system.

A conventional navigation system having a voice recognition function will be described with reference to FIGS. 3, 4 and 5. FIG.

FIG. 3 is a schematic configuration diagram of a conventional navigation system having a voice recognition function, FIG. 4 is a schematic configuration diagram of a voice recognition unit, and FIG. 5 is a diagram showing a transition state of a display unit in the navigation system. .

In FIG. 3, reference numeral 1 denotes a speech recognition unit for recognizing an inputted speech and outputting a recognition result, and 2 a speech for permitting speech input to the speech recognition unit 1 based on a trigger switch (not explicitly shown). The input designation unit 3 displays a list of recognition target words to be recognized in the speech recognition unit 1, a display unit that performs display based on the recognition result in the speech recognition unit 1, and the display unit 4 displays it on the display unit 3. A display unit 3 and a CD-ROM storing information;
Reference numeral 4 denotes a device to be controlled by the voice recognition unit 1. Reference numeral 5 denotes an antenna for receiving a signal from an external satellite, 6 denotes a GPS receiver for determining a current position of a vehicle or the like on which a navigation system is mounted based on the signal received by the antenna 5, and 7 denotes a CD-ROM. And the current display state of the navigation system based on the information of the GPS receiver 6 and
A state detection unit that inputs the current display state to the voice recognition unit 1;
Reference numeral 8 denotes a recognition target list table in which a recognition list for voice recognition displayed on the display unit 3 is stored according to the display state of the state detection unit 7, and 9 is a control unit that controls the entire system.

On the other hand, reference numeral 11 in FIG. 4 denotes a microphone for converting the input voice into an electric signal, 12 a voice analysis unit for analyzing the voice input to the microphone 11 and extracting characteristic parameters, and 13 a voice analysis unit. A section detection unit that detects a speech section based on the feature parameters extracted in 12, a pattern creation unit that creates a speech pattern based on the feature parameters of the section detected by the section detection unit 13, and 15 is an identification unit. The voice memory 16 stores a reference pattern used in advance in the voice memory, and 16 performs pattern matching between the reference pattern stored in the voice memory 15 and the voice pattern created by the pattern creation unit 14.
An identification unit that extracts the most similar reference pattern.

FIG. 5A shows an example of a display unit 3 when a current position is displayed in a vehicle equipped with a conventional navigation system (hereinafter referred to as a state). FIG.
(B) is an example of the display unit 3 when a destination is displayed in a vehicle equipped with a conventional navigation system (hereinafter, referred to as a state).

Before the user leaves for the destination, the user registers the destination in the navigation system in advance. For example, on the display unit 3, a map of Kyoto is selected, and the area around Kyoto Station is designated as the destination.

In the state (see FIG. 5A), the user presses the trigger switch (not explicitly shown) of the voice input designation unit 2 when he / she wants to search the map of the destination. As a result, when detecting that such a trigger switch has been pressed, the voice input designation unit 2 transmits the fact to the control unit 9 as a voice input permission signal. When receiving the voice input permission signal, the control unit 9 detects the current display state of the display unit of the navigation system detected by the state detection unit 7, and determines from the recognition target list table 8 that the state is necessary. A search target list, for example, List 1 in Table 1, is searched, and the recognition target vocabulary indicated in the list is transmitted to the speech recognition unit 1.

[0011]

[Table 1]

Thereafter, the user speaks a desired word, for example, <destination> to the microphone 11 of the voice recognition unit 1.
And say The speech analysis unit 12 extracts feature parameters from the speech <destination> input to the microphone 11. For example, bandpass filter analysis is performed by a filter bank, and the output of each bandpass filter is converted into a digital value by an analog / digital converter (not explicitly shown) in the voice analysis unit 12, and this value is used as a feature parameter as a section detection unit. 13 is transmitted.

The section detecting section 13 receives the characteristic parameters transmitted from the voice analyzing section 12, compares the characteristic parameters with a preset threshold value, and detects the start and end of the voice section. The pattern creation unit 14 patterns the feature parameters from the start to the end of the voice detected by the section detection unit 13. The voice pattern created by the pattern creation unit 14 is transmitted to the identification unit 16, and the identification unit 16
Pattern matching is sequentially performed with the reference patterns stored in the audio memory 15. Here, pattern matching is performed in order from number increase to number reduction, and the number of the most similar word <destination> is selected and output to the control unit 9.

The number output from the voice recognition unit 1 is transmitted to the control unit 9. The control unit 9 performs a process corresponding to the recognition result of the voice recognition unit 1. Since the processing in this case is to display the information of the destination on the display unit 3, the control unit 9
With reference to the destination code stored in the memory (not explicitly shown), the destination information is searched from the CD-ROM 4 and the destination information is displayed on the display unit 3. As a result, the display on the display unit 3 is changed as shown in the state (see FIG. 5B).

[0015]

However, in a navigation system having a voice recognition function, the display state of the navigation system is changed according to the result of the voice recognition unit 1, so that the display state is subject to voice recognition. The vocabulary is not always constant, but generally changes sequentially.

In this case, if the display unit 3 does not display any vocabulary to be currently subjected to speech recognition, the user may utter a word that is not the vocabulary to be recognized. In such a case, the speech recognition unit 1 selects the most similar word from the recognition target words, so that the navigation system malfunctions.

For this reason, a list of words currently recognized by the voice recognition unit 1 is displayed on the display unit 3 of the navigation system. According to this method, the display information required by the navigation system is displayed on the display
However, since the list of words to be recognized is redundantly displayed, a part of the display information required by the navigation system is missing, and there is a problem that desired display information cannot be obtained.

[0018]

SUMMARY OF THE INVENTION Accordingly, a speech recognition apparatus according to the present invention includes a speech input designating section for permitting a control section to input speech for controlling a device, a speech recognition section, A display unit for displaying at least one of desired display information obtained from a recognition result of the voice recognition unit, and a timer for measuring a time for displaying a recognition target list on the display unit. By this, when the voice input designation unit is in a voice input enabled state, the timer starts counting, and the recognition target list is displayed on the display unit, and when the timer reaches a predetermined time, the display information of the display unit is changed from the recognition target list. It is possible to switch to desired display information output by the navigation system.

[0019]

According to the above-mentioned means, when the voice input designation section is set to the voice input permission state, the timer starts measuring time and the recognition target list is displayed on the display section, while when the timer reaches a predetermined time. The recognition target list displayed on the display unit is deleted.

[0020]

FIG. 1 is a schematic block diagram showing an embodiment of a speech recognition apparatus according to the present invention.

In FIG. 1, components having the same functions as those of the conventional example shown in FIG. 3 are denoted by the same reference numerals, and description thereof will be omitted.

The difference between the present invention and the prior art is that the timer 10
The timer 10 and the voice input designating unit 2 display the recognition target list on the display unit 3 only for a predetermined time, and the timer 10 allows the voice input designating unit 2 to perform voice input. Then, counting is started, and when the count value reaches a preset count value, the fact is transmitted to the control unit 9.

FIG. 2 shows a display unit 3 of the speech recognition apparatus according to the present invention.
3 shows a transition state diagram.

When a trigger switch (not explicitly shown) of the voice input designation section 2 is pressed, a voice input permission signal is sent to the control section 9.
Is transmitted to When receiving the voice input permission signal, the control unit 9 instructs the timer 10 to start counting, and
The display state of the display unit 3 is read from the state detection unit 7. Since the current display state is the state, the list 1 of Table 1 required in that state is searched from the recognition target list table 8. The control unit 9 transfers the recognition target list to the display unit 3 according to the contents of the searched recognition target list, so that the recognition target list is displayed on the display unit 3 as shown in FIG.

Further, the control unit 9 instructs the timer 10 to start counting. When the timer 10 starts counting, the timer 10 continues counting up to a preset count value (for example, 10 seconds). When the count value reaches the preset count value, the timer 10 transmits a count end signal to the control unit 9. . The control unit 9 receives the count end signal from the timer 10, deletes the display of the recognition target list from the display unit 3, and displays the desired display information, which is the original display screen, which is output by the navigation system in FIG. Change and display as shown.

On the other hand, when the trigger switch (not explicitly shown) of the voice input designation unit 2 is pressed, the control unit 9 controls the voice recognition unit 1.
To the user that voice recognition is possible.

The voice recognition unit 1 recognizes voice input to the microphone 11 and transmits a recognition result to the control unit 9. The operation of the voice recognition unit 1 is the same as that of the conventional example, and the description is omitted here.

The control unit 9 outputs, for example, <
When the destination> is obtained, the internal memory of the control unit 9 (not explicitly shown)
Refer to the destination code stored in the CD-R
The information of the destination is searched from the OM 4 and the information of the destination is displayed on the display unit 3. As a result, the state (see FIG. 2C)
The display on the display unit 3 is changed as shown in FIG.

Thereafter, when inputting voice, the trigger switch of the voice input designating section 2 is pressed. When the trigger switch of the voice input designation unit 2 is pressed, a voice input permission signal is transmitted to the control unit 9. When receiving the voice input permission signal, the control unit 9 instructs the timer 10 to start counting and reads the display state from the state detection unit 7. Since the current display state is the state, the list 2 of Table 1 required in that state is searched from the recognition target list table 8 and displayed on the display unit 3. This state is shown in FIG.
(D).

With the above operation, when the voice input is specified in the voice input specifying unit 2, the recognition target list can be displayed on the display unit 3 for a certain period of time.

In this embodiment, the input by the voice input specifying unit 2 is specified by a manual switching input method. However, the present invention is not limited to the manual switching input. It is also possible to specify voice input in the voice input specifying unit 2 based on the recognition result in 1.

In this embodiment, the navigation system having the voice recognition function has been described. However, the contents of the present invention are not limited to the controlled equipment controlled by the voice recognition result such as the navigation system. It can be used for a voice recognition device having a display function.

In the present embodiment, the display state of the display unit 3 is limited to only two types of state and state. However, in addition to this, three or more display states including <transit point> and the like are described. It is needless to say that the recognition target list corresponding to each display state can be displayed together with the display state of the display unit 3 in this case.

[0034]

As is apparent from the above description, according to the present invention, when a voice input is specified in the voice input specifying unit, the timer counts the time for displaying the recognition target list on the display unit, so that a certain time is required. During the period, the recognition target list can be displayed.

Since the user speaks with reference to the recognition target list, it is possible to prevent words other than the recognition target from being uttered. Further, while the recognition target list is not displayed on the display unit, desired display information can be displayed, and it is possible to prevent a part of the desired display information from being lost. Can be obtained at a desired time.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of an embodiment of a navigation system having a voice recognition function according to the present invention.

FIG. 2 is a transition state diagram of a display unit of a navigation system having a voice recognition function according to the present invention.

FIG. 3 is a schematic configuration diagram of one embodiment of a conventional navigation system having a voice recognition function.

FIG. 4 is a schematic configuration diagram of an embodiment of a conventional voice recognition unit.

FIG. 5 is a transition state diagram of a display unit of a navigation system having a conventional voice recognition function.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Voice recognition part 2 Voice input designation part 3 Display part 7 State detection part 8 Recognition target list table 9 Control part 10 Timer

Continuation of front page (56) References JP-A-3-222002 (JP, A) JP-A-57-63599 (JP, A) JP-A-4-152399 (JP, A) JP-A-3-106276 (JP) JP-A-5-110908 (JP, A) JP-A-1-240922 (JP, A) JP-A-4-250489 (JP, A) JP-A-2-224100 (JP, A) 3-22491 (JP, U) JP 4-15479 (JP, B2) International Publication 91/8567 (WO, A1) (58) Fields investigated (Int. Cl. ⁷ , DB name) G10L 15/00- 15/28 G09B 29/10

Claims (3)

(57) [Claims] 1. A voice recognition unit for recognizing an input voice and outputting a recognition result, and a navigation system which receives an output of the voice recognition unit and is controlled by the voice recognition unit.
A control unit that controls a device for navigation , a recognition target list table storing a recognition target list of vocabulary to be recognized by the voice recognition unit, and the control unit controls the navigation system device for the voice recognition unit. A voice input designation unit for permitting input of voice for control, and at least one of desired display information for a navigation system device obtained from the recognition target list or a recognition result of the voice recognition unit. And a timer for measuring the time during which the recognition target list is displayed on the display unit. When the voice input specification unit specifies that voice input is possible, the timer operates. And displaying the recognition target list on the display unit on the display of desired display information for the navigation system device. On the other hand, when the timer reaches a predetermined time, the recognition target list displayed on the display unit is deleted. 2. The navigation system according to claim 1, wherein said means for permitting speech input by said speech input designation unit is a recognition result of said speech recognition unit.
System voice recognition device . 3. The key input device according to claim 1, wherein the input unit for permitting the voice input by the voice input designating unit is a key.
A speech recognition device for a navigation system according to the above.