JP2004184715A - Speech recognition apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a speech recognition apparatus with a high speech recognition rate. <P>SOLUTION: A passenger detection part 15 detects whether or not there is a passenger and when there is a passenger, the maximum wait time is shortened to complete speech input earlier than when there is no passenger, thereby decreasing the rate at which the voice of the passenger is included in the speech by a signal processing part 3 even if the passenger speaks while a user who is inputting a voice speaks. Consequently, deterioration in the speech recognition rate of the user by the signal processing part 3 can be reduced. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a voice recognition device that recognizes voice.
[0002]
[Prior art]
[Patent Document 1] Japanese Patent Application Laid-Open No. 2001-166794 Conventionally, there is a speech recognition device as described in Japanese Patent Application Laid-Open No. 2001-166794. When this voice recognition device is applied to an in-vehicle navigation system, it is determined whether or not there is only one occupant before voice recognition. If there are two or more occupants, attention is given to a non-user, ie, a passenger. By evoking, an environment in which only the user of the voice recognition device speaks is created in the vehicle, and the voice recognition rate is increased and the usability is improved.
[0003]
[Problems to be solved by the invention]
In the above-described conventional voice recognition device, when a non-user of the voice recognition device, that is, a passenger, misses the alert, or does not understand the meaning even after hearing the alert, etc. However, there is a problem that the fellow passenger utters during the voice recognition process, that is, before the user's utterance ends, thereby lowering the voice recognition rate. There is also a problem that the speech recognition rate is reduced when an environment suitable for speech recognition cannot be maintained due to a cause other than a passenger.
[0004]
In view of the above problems, an object of the present invention is to provide a speech recognition device that can perform speech recognition with higher accuracy.
[0005]
[Means for Solving the Problems]
The present invention provides a maximum standby time setting means for setting a maximum standby time from the start of audio capture to the end of the audio capture enabled state, and an end of the maximum standby time set by the maximum standby time setting means from the start of audio capture. In a speech recognition device having a signal processing unit for recognizing a voice input before, when there is a possibility that a sound other than the voice to be recognized may be captured, the maximum standby time setting means sets the maximum standby time to Was shortened.
[0006]
【The invention's effect】
According to the present invention, when there is a possibility that a sound other than the speech to be recognized may be captured, the maximum standby time is reduced so that the voice other than the voice to be recognized is included in the voice captured by the signal processing unit. Sound can be reduced in proportion. Therefore, deterioration of the voice recognition rate can be reduced.
[0007]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, embodiments of the present invention will be described with reference to examples.
In each of the embodiments described below, the speech recognition device according to the present invention is applied to a vehicle navigation system.
FIG. 1 shows the overall configuration of a vehicle navigation system according to the first embodiment.
A navigation control unit 2 that calculates the position of the vehicle based on a signal received by a GPS (Global Positioning System) antenna (not shown) and presents various information to a user is connected to the signal processing unit 3.
[0008]
The signal processing unit 3 includes a CPU, a memory, and the like, and performs a voice recognition process. The signal processing unit 3 is connected to a storage unit 6 that stores a grammar (described later) having a hierarchical structure of words to be subjected to speech recognition. The signal processing unit 3 is connected to an input unit 12 having an utterance switch 13 and a correction switch 14, and a passenger detection unit 15 for detecting the presence or absence of a passenger.
[0009]
Further, a speaker 9 is connected to the signal processing unit 3 via a D / A converter 7 and an output amplifier 8, and the digital audio signal output from the signal processing unit 3 is converted into an analog audio signal by the D / A converter 7. It is converted, amplified by the output amplifier 8 and output from the speaker 9 as sound.
A microphone 11 is connected to the signal processing unit 3 via an A / D converter 10, and an analog audio signal input from the microphone 11 is converted into a digital audio signal by the A / D converter 10, and the signal processing unit 3 Is transmitted to.
[0010]
The navigation control unit 2 is connected to the display unit 16 and the speaker 9, and presents position information and the like to a vehicle driver or the like through the display unit 16 and the speaker 9. The speech recognition unit 1 is composed of the signal processing unit 3, the storage unit 6, the D / A converter 7, the output amplifier 8, and the A / D converter 10.
The navigation system 20 includes the voice recognition unit 1, the navigation control unit 2, the display unit 16, the speaker 9, the microphone 11, the input unit 12, and the passenger detection unit 15.
[0011]
Next, the flow of the voice recognition processing of the navigation system in the present embodiment will be described using the flowchart of FIG.
In step 100, the signal processing unit 3 determines whether the user of the navigation system 20 has operated the utterance switch 13 provided on the input unit 12 for instructing the start of utterance. When the utterance switch 13 is operated, the process proceeds to step 101.
[0012]
In step 101, the signal processing unit 3 sets a grammar having a hierarchical structure of words stored in the storage unit 6 as a recognition target word. Here, the grammar having a hierarchical structure of words refers to an array of place names, and FIG. 3 shows an example thereof. First, a prefecture name is set at the beginning of a sentence, and then a place name is sequentially set, such as a municipal name corresponding to each prefecture.
[0013]
In step 102, the signal processing unit 3 sets the maximum standby time based on the recognition target word set in step 101. This maximum standby time is provided in order to avoid continuing the standby state without detecting a sound. The length of the maximum standby time Tm is obtained by adding the margin time Tx to the estimated utterance time Tw when the longest sentence included in the set recognition target word is uttered, and is expressed by the following equation.
Tm = Tw + Tx (1)
The allowance time Tx is usually about the same as the estimated utterance time Tw in order to absorb delays in utterance start and utterance length variations caused by individual differences if the user is unfamiliar with the navigation system 20 with the voice recognition device. Is set.
[0014]
In step 103, the signal processing unit 3 detects the presence or absence of a passenger by using the passenger detection unit 15. The passenger detection unit 15 detects the presence or absence of a passenger by using a pressure-sensitive sensor attached to a seat. When a passenger is detected, the process proceeds to step 104, and when a passenger is not detected, the process proceeds to step 105.
[0015]
In step 104, the signal processing unit 3 changes the maximum standby time Tm calculated by the equation (1) to a shorter value, and calculates the maximum standby time Tm 'as described below.
As shown in FIG. 4, when the utterance switch 13 is operated at time A and the voice capturing by the signal processing unit 3 is started, the utterance of the user usually occurs before the time C when the maximum standby time Tm of the voice ends. The processing ends at time B.
[0016]
However, as shown in FIG. 5, at time D before time B, at which the user's speech ends, if the fellow passenger's speech E is present halfway, the voice capturing process is greatly affected. As described above, the signal processing unit 3 cannot detect the end of the utterance of the user, and continues to utter the fellow passenger from the middle of the utterance of the user to capture the voice.
[0017]
As a result, voice capture is continued until time C when the maximum standby time Tm set in step 102 ends. Therefore, the captured voice includes the voice of the passenger in a wide range, and the recognition rate of the voice of the user deteriorates. To prevent this, the maximum standby time Tm 'is calculated using the following equation.
Tm '= Tw + Tx' (2)
Here, Tx ′ = k × Tx, and 0 <k <1.0.
[0018]
As a result, as shown in FIG. 6, even if there is an utterance of a passenger from time D, voice capture ends at time C ′ at which the maximum standby time Tm ′ ends. The range in which the utterance of the fellow passenger is included can be reduced.
[0019]
Returning to the flowchart of FIG. 2, in step 105, the signal processing unit 3 converts the notification sound stored in the storage unit 6 into the D / A converter 7 and the output amplifier 8 in order to notify the user of the start of the sound capturing process. Through the speaker 9.
[0020]
The user who hears the notification voice that notifies the start of voice capture utters a word included in the recognition target word. In this embodiment, the recognition target is the address shown in FIG.
The audio signal input from the microphone 11 is converted into a digital signal by the A / D converter 10 and input to the signal processing unit 3.
[0021]
Until the utterance switch 13 is operated, the signal processing unit 3 calculates the average power of the audio digital signal converted by the A / D converter 10. After the utterance switch 13 is operated, when the instantaneous power of the digital signal becomes larger than the calculated average power by a predetermined value or more, in step 106, it is determined that the user has uttered, and the voice is captured. Start.
[0022]
When the voice capture is started, the signal processing unit 3 starts calculating the degree of coincidence with the recognition target word stored in the storage unit 6 in step 107. The degree of coincidence indicates how similar the captured voice part is to the individual recognition target words, and the degree of coincidence is obtained as a score. In this embodiment, it is assumed that the higher the score value, the higher the matching degree.
It should be noted that during the processing of this step, the voice capturing by the signal processing unit 3 is continued in parallel.
[0023]
In step 108, it is determined whether the end of the utterance has been detected. The detection of the termination is to judge that the utterance of the user has ended when the instantaneous power of the audio digital signal becomes equal to or more than a predetermined time and equal to or less than a predetermined value. If the end of the utterance is detected, the process proceeds to step 109, and if not, the process proceeds to step 113.
[0024]
In step 113, it is determined whether or not the maximum standby time Tm or the maximum standby time Tm 'has elapsed after the start of voice capture. If not, the process returns to step 106. If the maximum standby time has elapsed, the process proceeds to step 109.
[0025]
In step 109, the voice capturing process ends, and in step 110, the signal processing unit 3 outputs the recognition target word having the highest degree of coincidence as the recognition result from the speaker 9 through the D / A converter 7 and the output amplifier 8. In the present embodiment, “Natsushima-cho, Yokosuka-shi, Kanagawa” spoken by the user is correctly recognized, and the signal processing unit 3 outputs the recognition result “Natsushima-cho, Yokosuka-shi, Kanagawa” through the speaker 9.
[0026]
In step 111, after outputting the recognition result in step 110, it is determined whether the correction switch 14 provided in the input unit 12 has been operated within a predetermined time. If the correction switch 14 has been operated, the user determines that the voice recognition result of the navigation system 20 has been corrected, returns to step 101, and performs the above-described voice capture again.
[0027]
On the other hand, if the correction switch 14 has not been operated within the predetermined time, the process proceeds to step 112, where it is determined that the user has accepted the recognition result of the navigation system 20, and a process according to the recognition result is performed. In this embodiment, the signal processing unit 3 outputs the address as the recognition result to the navigation control unit 2. The navigation control unit 2 sets the recognized address as the destination, and presents information such as road guidance to the user through the display unit 16 and the speaker 9.
In this embodiment, steps 102 and 104 in FIG. 2 constitute the maximum standby time setting means in the present invention. Further, the passenger detection unit 15 in the present embodiment constitutes a passenger detection unit in the present invention.
[0028]
The present embodiment is configured as described above, and when a passenger is detected by the passenger detection unit 15, the maximum standby time Tm is reduced to calculate the maximum standby time Tm ′, and the voice capturing time width is reduced. By doing so, even if the utterance of the fellow passenger occurs before the utterance of the user ends, the proportion of the voice of the fellow passenger in the voice captured by the signal processing unit 3 can be reduced. Therefore, it is possible to reduce the deterioration of the user's voice recognition rate by the signal processing unit 3.
[0029]
In this embodiment, the passenger detection unit 15 detects the presence or absence of a passenger by using a pressure-sensitive sensor attached to the seat. However, the present invention is not limited to this. The presence or absence of a passenger may be detected using an infrared sensor, an ultrasonic sensor, or the like that is blocked when sitting. In addition, a seatbelt wearing sensor may be used to detect the presence or absence of a passenger based on whether or not a seatbelt is worn outside the driver's seat. You may make it detect.
[0030]
Next, a second embodiment will be described.
FIG. 7 shows the overall configuration of a vehicle navigation system according to the second embodiment.
In the configuration of the present embodiment, the passenger detection unit 15 in the first embodiment is deleted, and a noise holding unit 18 is added.
A signal processing unit 3A including a CPU and a memory is connected to the navigation control unit 2 that calculates the position of the vehicle.
[0031]
The signal processing unit 3 </ b> A includes a noise holding unit 18 that holds noise information such as a sound or a sound close to the sound.
The speech recognition unit 1A is composed of the signal processing unit 3A, the storage unit 6, the D / A converter 7, the output amplifier 8, and the A / D converter 10.
A navigation system 20A is constituted by the voice recognition unit 1A, the navigation control unit 2, the display unit 16, the speaker 9, the microphone 11, and the input unit 12.
The components having the same functions as those of the first embodiment are denoted by the same reference numerals, and description thereof is omitted.
[0032]
Next, a flow of the voice recognition processing of the navigation system in the present embodiment will be described.
When the vehicle is traveling in an environment where noise occurs intermittently, or when the user and the passenger are talking, the signal processing unit 3A starts capturing voice as shown in FIG. Before time A, the sound or a sound close thereto is detected as noise G.
In such a case, there is a high possibility that a sound or a sound close thereto is detected as the noise H even after the start of the sound capturing.
[0033]
With reference to the flowchart of FIG. 9, a description will be given of a process of detecting the noise G before the time A at which the voice capturing starts.
This detection processing is periodically executed by interruption processing of a timer (not shown) incorporated in the signal processing unit 3A.
In step 200, the signal processing unit 3A determines whether or not the start point of noise has been detected within a predetermined time. In the detection of the starting point of the noise, the average power of the digital signal of the noise was calculated, and the noise was detected when the instantaneous power became larger than the average power by a predetermined value, similarly to step 106 in the first embodiment. Is determined.
[0034]
When the start point of the noise is detected in step 200, the signal processing unit 3A starts capturing noise in step 201. In step 202, when the end of the noise is detected, the acquisition of the noise ends. The detection of the end of the noise is performed in the same manner as in step 108 in the first embodiment.
[0035]
In step 203, the signal processing unit 3A stores the time at which the noise was taken and the duration thereof as noise data in the noise holding unit 18.
If the start point of the noise cannot be detected in step 200, the process is terminated.
[0036]
Next, the flow of voice recognition performed by the navigation system 20A will be described using the flowchart of FIG.
Steps 300 to 302 are the same as steps 100 to 102 in the first embodiment, and steps 304 to 313 in FIG. 10 are the same as steps 104 to 113 in the first embodiment, and a description thereof will be omitted.
[0037]
In step 303, the signal processing unit 3A determines whether noise has been detected before the user's voice input is started. This determination is made by using the noise data stored by the noise detection processing shown in the flowchart of FIG. 9 and integrating the duration of the noise that occurred between the time when the speech switch 13 was operated and a predetermined time before, If the integrated value is equal to or greater than a predetermined value, it is determined that noise has been detected. If noise is detected, the process proceeds to step 304, where the maximum standby time Tm is reduced. On the other hand, if no noise is detected, the process proceeds to step 305.
In this embodiment, steps 302 and 304 in FIG. 10 constitute the maximum standby time setting means in the present invention.
[0038]
This embodiment is configured as described above. If the signal processing unit 3A detects noise before the user starts inputting voice to the navigation system 20A, the maximum standby time Tm is set as shown in FIG. By shortening the maximum standby time Tm 'and ending the voice capture at the time C', even if there is a noise input other than the user's utterance, the voice capture by the signal processing unit 3A is performed. Can be reduced by the ratio of noise. Therefore, it is possible to reduce the deterioration of the user's voice recognition rate by the signal processing unit 3A.
[0039]
Next, a third embodiment will be described.
FIG. 11 shows an overall configuration of a vehicle navigation system according to the third embodiment.
In the configuration of the present embodiment, the passenger detection unit 15 in the first embodiment is deleted, a rear side monitoring unit 17 that monitors the rear and side of the vehicle is added, and the detection of the own vehicle position is further performed. This is an addition of the own vehicle position detection unit 19 and the environmental sound increase prediction unit 23 that perform the operation.
The navigation control unit 2B includes a host vehicle position detection unit 19 that calculates the position of the host vehicle from a signal from a GPS antenna (not shown). The navigation control unit 2B is connected to a signal processing unit 3B including a CPU and a memory.
[0040]
A rear side monitoring unit 17 that monitors the rear side of the vehicle is connected to the signal processing unit 3B. The signal processing unit 3B includes an environmental sound increase prediction unit 23 for predicting an increase in environmental sound.
The voice recognition unit 1B is composed of the signal processing unit 3B, the storage unit 6, the D / A converter 7, the output amplifier 8, and the A / D converter 10.
[0041]
The navigation system 20B includes the voice recognition unit 1B, the navigation control unit 2B, the display unit 16, the speaker 9, the microphone 11, the input unit 12, and the rear side monitoring unit 17.
The components having the same functions as those of the first embodiment are denoted by the same reference numerals, and description thereof is omitted.
[0042]
Next, the flow of the voice recognition process performed by the navigation system 20B will be described using the flowchart of FIG.
Steps 400 to 402 are the same as steps 100 to 102 in the first embodiment, and steps 404 to 413 are the same as steps 104 to 113 in the first embodiment, and a description thereof will be omitted.
[0043]
In step 403, a process of predicting whether or not the environmental sound increases is performed. The environmental sound includes an environmental sound accompanying an appearance of a passing vehicle or the like and an environmental sound generated by entering a tunnel.
The increase of the environmental sound accompanying the appearance of the overtaking vehicle can be predicted by the rear side monitoring unit 17 that monitors the rear side of the vehicle as an increase in the environmental sound when the vehicle approaching the own vehicle is detected. it can.
[0044]
As the rear side monitoring unit 17 for monitoring the rear side of the vehicle, a CCD camera can be used. As a method of monitoring the rear side of a vehicle using such a CCD camera, there is, for example, a rear side monitoring device for a vehicle disclosed in JP-A-2000-259998. In addition, other than the CCD camera, a rear-side radar for vehicle use may be used.
[0045]
In addition, the increase of the environmental sound caused by the intrusion into the tunnel is based on the position information of the own vehicle obtained by the own vehicle position detection unit 19 of the navigation control unit 2, so that the own vehicle can increase the environmental sound. A prediction can be made by determining whether or not the user has invaded. A method of detecting entry into a tunnel using such a navigation device is described in detail in, for example, a navigation device and accessory control method disclosed in Japanese Patent Application Laid-Open No. 2002-236023.
[0046]
In step 403, if an increase in environmental sound is expected, the process proceeds to step 404, in which the maximum standby time Tm is reduced, and the maximum standby time Tm 'is calculated. On the other hand, when it is not expected that the environmental sound increases, the process proceeds to step 405.
In this embodiment, the rear side monitoring unit 17 forms a vehicle monitoring unit according to the present invention. Steps 402 and 404 in FIG. 12 constitute the maximum standby time setting means in the present invention.
[0047]
The present embodiment is configured as described above, and predicts that the environmental sound will increase when the own vehicle enters the tunnel or when a vehicle approaching the own vehicle is detected by the rear side monitoring unit 17. To reduce the maximum standby time Tm. As a result, as shown in FIG. 13, the capture of the voice starts at time A, the environmental sound I is generated during the voice utterance of the user, and even if the utterance of the user overlaps with the environmental sound I, By ending the audio capture at the time C ′ at which the maximum standby time Tm ′ ends, the ratio of the environmental sound I to the audio captured by the signal processing unit 3B can be reduced. Therefore, it is possible to reduce the deterioration of the user's voice recognition rate by the signal processing unit 3B.
In the present embodiment, the rear side monitoring unit 17 is used to monitor the rear side of the vehicle. However, the present invention is not limited to this, and the surroundings of the vehicle such as the front side of the vehicle may be monitored.
[0048]
Next, a fourth embodiment will be described.
FIG. 14 shows the overall configuration of a vehicle navigation system according to the fourth embodiment.
The overall configuration of the present embodiment is obtained by deleting the fellow passenger detection unit 15 in the first embodiment and adding a use experience storage unit 21 that counts the number of executions of the voice recognition process. The signal processing unit 3C including a CPU and a memory includes a use experience storage unit 21 that counts the number of executions of the voice recognition process.
[0049]
The speech recognition unit 1C is configured by the signal processing unit 3C, the storage unit 6, the D / A converter 7, the output amplifier 8, and the A / D converter 10.
A navigation system 20C includes the voice recognition unit 1C, the navigation control unit 2, the display unit 16, the speaker 9, the microphone 11, and the input unit 12. Other configurations and operations are the same as those of the first embodiment, and the same reference numerals are given and the description is omitted.
[0050]
Next, the operation of the navigation system to which the voice recognition device according to the present embodiment is applied will be described.
The flow of the voice recognition process performed by the navigation system 20C will be described with reference to the flowchart of FIG.
Steps 500 to 502 are the same as steps 100 to 102 in the first embodiment, and steps 504 to 513 in FIG. 15 are the same as steps 104 to 113 in the first embodiment, and a description thereof will be omitted.
[0051]
In step 503, the signal processing unit 3C determines whether or not the user has sufficient experience using the navigation system to which the voice recognition device is applied. The signal processing unit 3C uses the execution count value of the voice recognition process counted by the use experience storage unit 21 to determine that the use experience is sufficient when the execution count value is equal to or greater than a predetermined value. It is.
[0052]
If it is determined in step 503 that the user has sufficient experience, the process proceeds to step 504 to reduce the maximum standby time Tm. On the other hand, if it is determined that the user experience is not sufficient, the process proceeds to step 505.
In this embodiment, steps 502 and 504 in FIG. 15 constitute the maximum standby time setting means in the present invention.
[0053]
The present embodiment is configured as described above, and the user recognizes that when the use experience of the voice recognition device increases, the voice recognition rate increases when the utterance starts immediately after the start of voice capture. Become. Therefore, compared to a user who has little experience using the voice recognition device, a user who has a lot of experience has a shorter end time of the utterance. By reducing the maximum standby time Tm in accordance with the reduction in the utterance time width, it is possible to reduce the proportion of voices other than the user occupying in the voices captured by the signal processing unit 3C. Therefore, it is possible to reduce the deterioration of the user's voice recognition rate by the signal processing unit 3C.
[Brief description of the drawings]
FIG. 1 is a diagram showing a first embodiment of the present invention.
FIG. 2 is a diagram showing a flow of a voice recognition process in the first embodiment.
FIG. 3 is a diagram showing a grammar having a hierarchical structure of words.
FIG. 4 is a diagram showing the relationship between the maximum standby time and speech.
FIG. 5 is a diagram showing the relationship between the maximum standby time and speech.
FIG. 6 is a diagram showing the relationship between the maximum standby time and speech.
FIG. 7 is a diagram showing a second embodiment.
FIG. 8 is a diagram showing the relationship between the maximum standby time and speech.
FIG. 9 is a diagram showing a flow of a noise capturing process.
FIG. 10 is a diagram showing a flow of a voice recognition process in the second embodiment.
FIG. 11 is a diagram showing a third embodiment.
FIG. 12 is a diagram showing a flow of a voice recognition process in the third embodiment.
FIG. 13 is a diagram showing the relationship between the maximum standby time and speech.
FIG. 14 is a diagram showing a fourth embodiment.
FIG. 15 is a diagram showing a flow of a voice recognition process in the fourth embodiment.
[Explanation of symbols]
1, 1A, 1B, 1C Voice recognition unit 2, 2B Navigation control unit 3, 3A, 3B, 3C Signal processing unit 6 Storage unit 7 D / A converter 8 Output amplifier 9 Speaker 10 A / D converter 11 Microphone 12 Input unit 13 Speech switch 14 Correction switch 15 Passenger detection unit 16 Display unit 17 Rear side monitoring unit 18 Noise holding unit 19 Own vehicle position detection unit 20, 20A, 20B, 20C Navigation system 21 Usage experience storage unit 23 Environmental sound increase prediction unit

Claims (7)

Maximum standby time setting means for setting a maximum standby time from the start of voice capture to the end of the voice capture enabled state,
A voice recognition device having a signal processing unit that recognizes voice input from the start of voice capture to the end of the maximum standby time set by the maximum standby time setting unit.
The speech recognition apparatus according to claim 1, wherein the maximum standby time setting means shortens the maximum standby time when a sound other than the voice to be recognized may be captured. Equipped with an attendee detection unit that detects the presence of an attendee,
The voice recognition device according to claim 1, wherein when the co-located person is detected by the co-located person detection unit, there is a possibility that a sound other than the voice to be recognized is taken in. A noise holding unit that holds a sound signal before the start of capturing the sound,
2. The speech recognition apparatus according to claim 1, wherein when the sound signal is equal to or more than a predetermined value, there is a possibility that a sound other than the speech to be recognized may be captured. An environmental sound increase prediction unit that predicts an increase in ambient environmental sound is provided,
2. The speech recognition apparatus according to claim 1, wherein a time when an increase in environmental sound is predicted by the environmental sound increase prediction unit is a case where a sound other than the voice to be recognized is likely to be captured. . Mounted on the vehicle,
A vehicle position detecting unit that detects a position of the vehicle,
The environmental sound increase prediction unit predicts that the environmental sound increases when the vehicle enters an area where the increase in the environmental sound is known, based on the position information detected by the host vehicle position detection unit. The speech recognition device according to claim 4, wherein: Mounted on the vehicle,
A vehicle monitoring unit that detects vehicles around the vehicle,
The voice recognition device according to claim 4, wherein the environmental sound increase predicting unit predicts that environmental sound increases when the vehicle is detected by the vehicle monitoring unit. Maximum standby time setting means for setting a maximum standby time from the start of voice capture to the end of the voice capture enabled state,
A voice recognition device having a signal processing unit that recognizes voice input from the start of voice capture to the end of the maximum standby time set by the maximum standby time setting unit.
A use experience storage unit that counts the number of executions of the voice recognition process by the signal processing unit,
The speech recognition apparatus according to claim 1, wherein the maximum standby time setting means shortens the maximum standby time when the coefficient value stored in the use experience storage unit is equal to or greater than a predetermined value.

Images