JP2017146531A - Voice guidance system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice guidance system capable of providing guidance with a voice under a condition that it is easily heard by a driver.SOLUTION: A voice guidance system comprises: a guidance part 33 for providing guidance with a voice; a parameter setting part 31 for setting an initial value of a parameter in accordance with a registered driver; a timing adjustment part 32 for adjusting a timing of providing guidance; a response detection part 34 for detecting a response of a driver; a sound volume adjustment part 35 and a speaking speed adjustment part 36 for adjusting a parameter on the basis of a detection result of the response detection part 34; and a parameter learning part 39 for learning an initial value of the parameter for each registered driver. When the guidance is unheard by the driver, the voice guidance system adjusts the parameter such that the sound volume of the voice becomes larger, or that the speaking speed becomes slower, and then provides the guidance again, and when the guidance is heard by the driver, causes the parameter learning part 39 to learn the initial value of the parameter.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a voice guidance system that provides various guidance to a driver by voice.

  Patent Document 1 discloses a robot that outputs voice again when the user's utterance after outputting voice indicates that the voice output by the robot cannot be heard as a robot that can interact with the user. . Patent Document 1 discloses that when the robot outputs the sound again, the user outputs a higher sound than the previous time or lowers the sound output speed than the previous time so that the user can easily hear the sound. .

JP 2008-168375 A

  By the way, in a car, for example, a driver who has a high driving load such as when turning right cannot afford to hear voice guidance and may not notice the voice guidance itself. For this reason, it is required to provide voice guidance under conditions that are easy for the driver to hear. However, simply applying the technique described in Patent Document 1 described above to an automobile cannot meet such a demand.

  An object of the present invention is to provide a voice guidance system capable of performing voice guidance under conditions that allow a driver to easily hear.

Hereinafter, means for solving the above-described problems and the effects thereof will be described.
A voice guidance system that solves the above problems includes a guidance unit that performs guidance to a driver of a vehicle by voice, and a parameter setting unit that sets, for each driver, initial values of parameters related to the guidance including the volume and the speech speed of the voice. A timing adjustment unit that determines whether or not the driver can easily hear the voice and adjusts the timing for performing the guidance; a response detection unit that detects a driver's response to the guidance; and the response The response includes a parameter adjustment unit that adjusts the parameter based on a detection result of the detection unit, and a parameter learning unit that learns an initial value of the parameter for each driver, and the response that the driver cannot hear the guidance. When the detection unit detects, the parameter adjustment unit is set so that the volume of the voice is increased or the speech speed is decreased. After adjusting the parameters, the guidance unit performs guidance again, and when the response detection unit detects that the driver has heard the guidance, the parameter value at that time and the initial value of the parameter Learning by the parameter learning unit is performed based on the comparison result with the value.

  According to the above configuration, by adjusting the timing at which the guidance is performed by the timing adjustment unit, it is possible to perform guidance by voice in a situation where the driver can easily hear in consideration of, for example, in-vehicle volume and driving load. When the response detection unit detects that the driver has not heard the guidance, the guidance is performed again by increasing the volume or reducing the speech speed so that the driver can hear more easily. When the response detection unit detects that the driver has heard the guidance, the driver learns the initial value of each parameter based on the comparison result between the value of each parameter and the initial value of each parameter. Done every time. As a result, the initial value of each parameter initially set in the voice guidance process approaches a value suitable for the driver. As a result, voice guidance can be performed under conditions that are easy for the driver to hear.

The functional block diagram which shows schematic structure of one Embodiment of a voice guidance system. The flowchart which shows an example of the procedure of a voice guidance process. The flowchart which shows an example of the procedure of a setting process. The flowchart which shows an example of the procedure of a timing adjustment process. The figure which shows an example of transition of the learning value of each parameter.

An embodiment of a voice guidance system will be described with reference to FIGS.
This voice guidance system identifies the driver when performing guidance by voice to the driver of the vehicle, and then sets initial values of voice-related parameters such as voice volume and speech speed, vehicle volume, and vehicle steering angle. The initial value of the parameter related to the timing for performing such guidance is set for each driver. Then, the voice guidance system provides guidance with timing and voice based on the initial values of the parameters. The voice guidance system detects a driver's response to voice guidance. When it is detected that the driver cannot hear the guidance, the voice guidance system adjusts the parameter values described above so that the driver can easily hear the voice, and then performs voice guidance again. On the other hand, when it is detected that the driver can hear the guidance, the voice guidance system determines the parameter to be set when performing the next guidance based on the comparison result between the parameter value at that time and the initial parameter value. Learn about initial values. The voice guidance system learns these parameters for each driver.

  As shown in FIG. 1, the voice guidance system is configured around a voice guidance device 30. The voice guidance device 30 is mainly configured by a microcomputer including a processing unit, a storage unit, an external input unit, an external output unit, and the like.

  The voice guidance device 30 is electrically connected to the input microphone 11. The input microphone 11 detects a sound in the vehicle such as a driver's speech, for example, converts the detected sound into a voice signal, and outputs the voice signal to the voice guidance device 30.

A blinker 12 is electrically connected to the voice guidance device 30. The turn signal 12 outputs an operation signal indicating the operation state of the turn signal 12 to the voice guidance device 30.
A vehicle speed sensor 13 is electrically connected to the voice guidance device 30. The vehicle speed sensor 13 detects the vehicle speed and outputs a vehicle speed signal indicating the detected vehicle speed to the voice guidance device 30.

  A shift position sensor 14 is electrically connected to the voice guidance device 30. The shift position sensor 14 detects the position of the shift lever mounted on the vehicle and outputs a position signal indicating the detected position to the voice guidance device 30.

  A steering angle sensor 15 is electrically connected to the voice guidance device 30. The steering angle sensor 15 detects the steering angle of the vehicle based on the steering angle of the steering, and outputs a steering angle signal indicating the detected steering angle to the voice guidance device 30. For example, the steering angle is indicated by a positive angle in the right direction in the front direction and a negative angle in the left direction with respect to the front direction, with the angle at which the vehicle travels straight as a reference angle.

  A driver monitor camera 16 is electrically connected to the voice guidance device 30. The driver monitor camera 16 includes the driver's face in the imaging region, and outputs image information indicating the captured imaging region to the voice guidance device 30.

  An input device 17 is electrically connected to the voice guidance device 30. The input device 17 is, for example, a touch panel device provided in a display constituting the navigation system, and outputs an operation signal indicating an operation by a driver or the like to the voice guidance device 30.

  A voice output speaker 18 is electrically connected to the voice guidance device 30. The voice guidance device 30 outputs a voice signal indicating guidance to the output speaker 18, and the output speaker 18 converts the voice signal output from the voice guidance device 30 into a voice and outputs the voice.

  A display device 19 is electrically connected to the voice guidance device 30. The display device 19 is a display that constitutes a navigation system, for example, and displays the image information output to the display device 19 by the voice guidance device 30 when performing voice guidance.

  The voice guidance device 30 includes a parameter setting unit 31, a timing adjustment unit 32, a guidance unit 33, a response detection unit 34, a volume adjustment unit 35, a speech speed adjustment unit 36, a rudder angle upper limit adjustment unit 37, and a noise upper limit adjustment unit 38. And a parameter learning unit 39, which executes voice guidance processing for providing guidance by voice.

  The parameter setting unit 31 executes a setting process for setting initial values of parameters related to voice for guidance and initial values of parameters related to timing when voice guidance is performed. The parameters set by the parameter setting unit 31 are a sound volume relative value and a speech speed that are parameters related to voice, and a steering angle upper limit value and a noise upper limit value that are parameters related to the timing of guidance. The parameter setting unit 31 identifies a driver based on, for example, an image recognition process for image information from the driver monitor camera 16 or an identification number of a vehicle key, and sets initial values of parameters according to the identified driver. .

  The timing adjustment unit 32 executes a timing adjustment process for adjusting the timing for performing guidance by voice. In the timing adjustment process, the timing adjustment unit 32 includes an audio signal from the input microphone 11, an operation signal from the winker 12, a vehicle speed signal from the vehicle speed sensor 13, a position signal from the shift position sensor 14, and a steering angle sensor 15. The timing for performing guidance is adjusted based on the steering angle signal, parameters relating to timing for performing guidance, and the like.

  The guide unit 33 provides voice guidance to the driver by outputting a voice signal for guidance to the output speaker 18. The guide unit 33 obtains the reference volume by adding the relative volume value to the in-vehicle volume based on the audio signal from the input microphone 11. When the reference volume is larger than a predetermined minimum volume (for example, “40 dB”) and smaller than a predetermined maximum volume (for example, “90 dB”), the guide unit 33 performs guidance with the reference volume. In addition, the guide unit 33 guides at the minimum volume when the reference volume is equal to or lower than the minimum volume, and guides at the maximum volume when the reference volume is equal to or higher than the maximum volume.

  The response detection unit 34 detects the driver's response to the guidance of the guide unit 33. When the guidance by the guidance unit 33 is performed, the response detection unit 34 outputs image information to that effect to the display device 19 when the driver needs to operate the input device 17. The response detection unit 34 receives a voice signal indicating the driver's speech from the input microphone 11 when a specific operation signal is input from the input device 17 within a predetermined time after outputting the image information. If it is, it is detected that the driver has heard the guidance. On the other hand, the response detection unit 34, when a specific operation signal is not input from the input device 17 within a predetermined time, and when a voice signal indicating the driver's speech is not input from the input microphone 11. Detect that the driver could not hear the guidance.

  The volume adjustment unit 35 is one of the parameter adjustment units, and adjusts the volume relative value, which is a parameter related to voice, when the response detection unit 34 detects that the driver has not heard voice guidance. When the response detection unit 34 detects that the driver has not heard the voice guidance, the volume adjustment unit 35 increases the volume relative value by “5 dB”, which is a unit adjustment amount of the volume relative value. Guidance is given by volume sound. In this volume relative value, a minimum relative value is set as a minimum value. In the present embodiment, “20 dB” is set as the minimum relative value, and this value is a value indicating a sound volume that is felt to be slightly larger than the vehicle sound volume.

  The speech speed adjustment unit 36 is one of parameter adjustment units, and adjusts the speech speed, which is a parameter related to speech, when the response detection unit 34 detects that the driver has not been able to hear voice guidance. The speaking speed is indicated by the number of mora that is the number of mora (beats) per second. When the response detection unit 34 detects that the driver cannot hear the voice guidance, the speech speed adjustment unit 36 decreases the mora number by “1 mora”, which is the unit adjustment amount of the mora number, and more slowly. Guidance is given with the tone of the tone. As the number of mora, the minimum number of mora that is the minimum value and the maximum number of mora that is the maximum value are set. In this embodiment, “4 mora” is set as the minimum number of mora and “8 mora” is set as the maximum number of mora, and this maximum number of mora is a value indicating an average speech speed.

  The steering angle upper limit adjustment unit 37 is one of the parameter adjustment units, and is a parameter related to the timing at which guidance is performed when the response detection unit 34 detects that the driver has not heard voice guidance. Adjust the rudder angle upper limit. The steering angle upper limit value is an upper limit value of the steering angle of the vehicle when guidance by the guide unit 33 is performed, and is a change amount of the steering angle with respect to the reference angle. The steering angle upper limit adjustment unit 37, when the response detection unit 34 detects that the driver has not been able to hear voice guidance, the steering angle upper limit value by “1 °” which is a unit adjustment amount of the steering angle upper limit value. The voice guidance is performed when the traveling direction of the vehicle is closer to the straight traveling direction. As the steering angle upper limit value, a minimum steering angle upper limit value that is a minimum value and a maximum steering angle upper limit value that is a maximum value are set. In this embodiment, “2 °” is set as the minimum steering angle upper limit value and “5 °” is set as the maximum steering angle upper limit value, and the maximum steering angle upper limit value is set so that the vehicle travels along a gentle curve. It is a value measured when

  The noise upper limit adjustment unit 38 is one of the parameter adjustment units, and is a noise related to timing at the time of guidance when the response detection unit 34 detects that the driver has not been able to hear voice guidance. Adjust the upper limit. The noise upper limit value is an upper limit value of the volume in the vehicle when guidance by the guide unit 33 is performed. The noise upper limit adjustment unit 38 reduces the noise upper limit by “5 dB” which is a unit adjustment amount of the noise upper limit when the response detection unit 34 detects that the driver cannot hear the voice guidance. Provide voice guidance when the interior of the car is quieter. In this noise upper limit value, a minimum noise upper limit value which is a minimum value and a maximum noise upper limit value which is a maximum value are set. In this embodiment, “40 dB” is set as the minimum noise upper limit value and “70 dB” is set as the maximum noise upper limit value, and this maximum noise upper limit value is measured when a considerably large noise is generated. Value.

  The parameter learning unit 39 learns, for each driver identified by the parameter setting unit 31, the initial value of each parameter that allows the driver to hear the voice stably. The parameter learning unit 39 holds, for each driver, learning data 40 that is data including a learning value that is a value of each parameter and an initial value of each parameter that is set by the parameter setting unit 31 in the next voice guidance process. Yes. This learning will be described later.

With reference to FIGS. 2 to 5, the voice guidance process executed by the voice guidance device 30 will be described.
As shown in FIG. 2, in the first step S <b> 101, the parameter setting unit 31 executes a setting process for setting each parameter related to the voice for guidance.

  As shown in FIG. 3, in the setting process, the parameter setting unit 31 identifies the driver in the first step S201, and whether or not the parameter learning unit 39 holds the learning data 40 of the identified driver. Judging. When the parameter learning unit 39 holds the driver's learning data 40 (step S201: YES), the parameter setting unit 31 sets initial values of parameters based on the learning data 40 in the next step S202, Return to the voice guidance process.

  On the other hand, when the parameter learning unit 39 does not hold the driver's learning data 40 (step S201: NO), the parameter setting unit 31 sets the initial value of each parameter in subsequent steps S203, S204, S205, and S206. Set the default value and return to the voice guidance process. In the present embodiment, the default value of the noise upper limit value is “70 dB” which is the maximum noise upper limit value, the default value of the steering angle upper limit value is “5 °” which is the maximum steering angle upper limit value, The default value is “20 dB” which is the minimum sound volume relative value, and the default value of the number of mora is “8 mora” which is the maximum number of mora.

  Returning to FIG. 2, in the next step S <b> 102, the timing adjustment unit 32 executes a timing adjustment process for measuring the timing for performing guidance by voice. Through the timing adjustment process, the timing adjustment unit 32 determines whether the driver can easily hear voice guidance for the surrounding situation of the vehicle and the driving situation of the vehicle.

  As shown in FIG. 4, in the timing adjustment process, the timing adjustment unit 32 measures the volume in the vehicle based on the audio signal from the input microphone 11 in the first step S301. In the next step S302, the timing adjustment unit 32 determines whether or not the in-vehicle volume measured in step S301 is less than the noise upper limit value. When the vehicle interior volume is equal to or higher than the noise upper limit value (step S302: NO), the timing adjustment unit 32 repeatedly executes step S301 and step S302 until the vehicle interior volume becomes less than the noise upper limit value. If the volume in the vehicle is less than the noise upper limit (step S302: YES), the timing adjustment unit 32 determines that the noise in the vehicle is easy for the driver to hear voice guidance, and proceeds to the next step S303. move on.

  In step S <b> 303, the timing adjustment unit 32 receives the operation signal from the turn signal 12, the vehicle speed from the vehicle speed sensor 13, the position signal from the shift position sensor 14, the steering angle signal from the steering angle sensor 15, and the driver monitor camera 16. Image information is acquired as vehicle information.

  The timing adjustment unit 32 that has acquired the vehicle information determines whether or not the turn signal 12 is in an off state (step S304). When the turn signal 12 is in an off state (step S304: YES), the timing adjustment unit 32 determines that the vehicle is not in a right / left turn state or a lane change state, and then whether or not the vehicle is in a departure / reverse state. Is determined (step S305). In step S305, the timing adjustment unit 32 determines whether or not the vehicle is in the departure / reverse state based on the vehicle speed signal from the vehicle speed sensor 13, the position signal from the shift position sensor 14, and the like.

  When the vehicle is not in the departure / reverse state (step S305: NO), the timing adjustment unit 32 determines whether or not the steering angle is less than the steering angle upper limit value (step S306). When the rudder angle is less than the rudder angle upper limit value (step S306: YES), the timing adjustment unit 32 performs image processing on the image information from the driver monitor camera 16 so that the driver's line of sight is the front direction of the vehicle. It is determined whether or not (step S307). When the driver's line of sight is the front direction of the vehicle (step S307: YES), the timing adjustment unit 32 determines that the driver's driving load is small and that the driver can easily hear guidance. Return to the voice guidance process.

  On the other hand, when the turn signal 12 is in the on state (step S304: NO), the timing adjustment unit 32 is in the departure / reverse state (step S305: YES), and the steering angle is greater than or equal to the steering angle upper limit value. (Step S306: NO) If the driver's line of sight is not in the front direction of the vehicle (Step S307: NO), the process returns to Step S301 and the in-vehicle volume is measured again.

  Returning to FIG. 2, in the next step S <b> 103, the guide unit 33 provides voice guidance to the driver. Further, the guide unit 33 outputs image information that requires a driver's operation to the input device 17 to the display device 19. In the next step S104, the response detection unit 34 detects the driver's response to the guidance of the guide unit 33 according to the presence or absence of a response from the driver. When the response detection unit 34 detects that the driver has heard the guidance (step S104: YES), the parameter learning unit 39 performs parameter learning (step S105), and then the voice guidance process ends.

  On the other hand, when the response detection unit 34 detects that the driver has not been heard (step S104: NO), in the next step S106, the volume adjustment unit 35 determines whether or not the guidance in step S103 is performed at the maximum volume. Determine whether. When the guidance is not performed at the maximum volume (step S106: NO), the volume adjustment unit 35 increases the volume relative value by “5 dB” which is the unit adjustment amount of the volume relative value (step S107). Then, after the timing adjustment process in step S102, guidance by the guide unit 33 is performed again in step S103.

  When the guidance is performed at the maximum volume (step S106: YES), the speech speed adjustment unit 36 determines whether the guidance at step S103 is performed with the minimum number of mora (step S108). When the guidance is not performed with the minimum number of mora (step S108: NO), the speech speed adjustment unit 36 reduces the speech speed by reducing the number of mora by “1 mora” which is the unit adjustment amount of the mora number. (Step S109). Then, after the timing adjustment process in step S102, guidance by the guide unit 33 is performed again in step S103.

  On the other hand, if the guidance is performed with the minimum number of mora (step S108: YES), then the rudder angle upper limit value adjustment unit 37 determines that the rudder angle upper limit value used in the timing adjustment process of step S102 is the smallest rudder value. It is determined whether or not the angle is the upper limit value. When the rudder angle upper limit value is not the minimum rudder angle upper limit value (step S110: NO), the rudder angle upper limit value adjustment unit 37 sets the rudder angle upper limit value by “1 °” which is a unit adjustment amount of the rudder angle upper limit value. The guidance is not performed unless the traveling direction of the vehicle is closer to the straight traveling direction (step S111). Then, after the timing adjustment process in step S102, guidance by the guide unit 33 is performed again in step S103.

  On the other hand, if the rudder angle upper limit value is the minimum rudder angle upper limit value (step S110: YES), then the noise upper limit value adjustment unit 38 determines that the noise upper limit value used in the timing adjustment process is the minimum noise upper limit value. It is determined whether or not. When the noise upper limit value is not the minimum noise upper limit value (step S112: NO), the noise upper limit value adjusting unit 38 reduces the noise upper limit value by “5 dB” which is the unit adjustment amount of the noise upper limit value, and is quieter. If there is no situation, guidance is not performed (step S113). Then, after the timing adjustment process in step S102, guidance by the guide unit 33 is performed again in step S103.

  On the other hand, when the noise upper limit is the minimum noise upper limit (step S112: YES), that is, the volume is the maximum volume, the number of mora is the minimum number of mora, the steering angle upper limit is the minimum steering angle upper limit, and the noise upper limit is If it is the minimum noise upper limit value, parameter learning by the parameter learning unit 39 (step S105) is performed, and the voice guidance process ends.

Next, an example of parameter learning by the parameter learning unit 39 will be described in detail with reference to FIG.
As described above, the parameter learning unit 39 learns for each driver the initial value of each parameter set by the parameter setting unit 31 in the setting process of step S101. The parameter learning unit 39 sets the initial value of each parameter based on the result of comparing the initial value of each parameter set by the parameter setting unit 31 with the response value that is the value of each parameter set at the time of step S105. Learn the value.

  If the parameter setting unit 31 determines that the learning data 40 of the driver currently driving is not held, the parameter learning unit 39 registers the driver by creating the learning data 40 corresponding to the driver. To do. The learning data 40 is data that is given an ID number for each driver, and indicates the learning value of each parameter and the initial value of each parameter set by the parameter setting unit 31 in the next voice guidance process. The parameter learning unit 39 acquires the response value of each parameter immediately after creating the learning data 40 as the first learning value. Then, the parameter learning unit 39 updates the learning value of each parameter by adjusting or confirming the parameter based on the learning value.

  As shown in FIG. 5, the parameter learning unit 39 selects and adjusts an adjustment parameter when at least one of the updated learned values is not a default value as in the first learned value. The parameter learning unit 39 selects one of the parameters whose learning value is not the default value as the adjustment parameter. The parameter learning unit 39 selects each parameter as an adjustment parameter in the order of relative volume value, number of mora, rudder angle upper limit value, and noise upper limit value. In the case of the first learning value, the parameter learning unit 39 selects the relative volume value as the adjustment parameter. Then, the parameter learning unit 39 sets a value that is close to the default value by the unit adjustment amount from the learning value for the adjustment parameter as an adjustment value, and sets the adjustment value as an initial value as the adjustment value for the remaining parameters. Set.

  In the next voice guidance process, the parameter setting unit 31 sets the adjustment value described above as the initial value of each parameter. Then, when the response value in the voice guidance process is the same value as the adjustment value, the parameter learning unit 39 sets the adjustment value as an initial value again as a confirmation value. The parameter learning unit 39 checks the adjustment value once more. The parameter learning unit 39 then learns by setting the adjustment value of the first learning value to the second learning value when the voice guidance process in which the response value becomes the same value as the adjustment value continues three times after the parameter adjustment. Update the value.

  The second learning value has a sound volume relative value and a noise upper limit value as parameters that are not default values. In this case, in principle, the parameter learning unit 39 selects the mora number that is the next order of the relative volume value as the adjustment parameter, but the mora number is the default value, and the steering angle upper limit value is also set. Since it is a default value, the noise upper limit value, which is the next order, is selected as the adjustment parameter. Then, after the parameter adjustment, the parameter learning unit 39 performs the second learning in which the noise upper limit value is brought close to the default value by the unit adjustment amount when the voice guidance process in which the response value becomes the same value as the adjustment value continues three times. The learning value is updated by setting the value adjustment value to the third learning value.

  The third learning value has a sound volume relative value as a parameter that is not a default value. In this case, the parameter learning unit 39 selects, as an adjustment parameter, a volume relative value that is the next parameter of the noise upper limit value according to the order. When the response value is different from the adjustment value in the next voice guidance process, the parameter learning unit 39 returns the adjustment parameter value to the learning value, and sets the learning value of each parameter as the adjustment value to the initial value. To do. The parameter learning unit 39 sets the third learning value to the fourth learning value when the response value of each parameter is the same as the adjustment value in the next voice guidance process in which the learning value is set as the adjustment value of each parameter. The learning value is updated by setting. In addition, when the response value of each parameter is different from the adjustment value in the next voice guidance process in which the learning value is set as the adjustment value of each parameter, the parameter learning unit 39 sets the response value to the fourth learning value. The learning value is updated by setting to.

  In this way, the parameter learning unit 39 learns the initial value of each parameter so that the first guidance in each voice guidance process can be performed under conditions that the driver can easily hear.

According to the voice guidance system of the above embodiment, the following effects can be obtained.
(1) By adjusting the timing at which the guidance is performed by the timing adjustment unit 32, it is possible to perform voice guidance in a situation where the driver can easily hear in consideration of the volume in the vehicle, the driving load of the driver, and the like. When the response detection unit 34 detects that the driver cannot hear the guidance, the guidance is performed again by increasing the volume or reducing the speech speed so that the driver can hear more easily. The parameter learning unit 39 learns for each driver the initial value of each parameter initially set in the voice guidance process based on the comparison result between the response value of each parameter and the initial value of each parameter. . As a result, voice guidance can be performed under conditions that are easy for the driver to hear.

(2) The volume of voice guidance is determined by using a volume obtained by adding a relative volume value to a vehicle volume as a reference volume. Thereby, the volume of guidance can be adjusted according to the noise in the vehicle.
(3) The volume inside the vehicle is less than the noise upper limit value, the blinker 12 is in the off state, the vehicle is not in the departure / reverse state, the driver's line of sight is in the front direction, and all of these are established. Voice guidance is provided when According to such a configuration, it is possible to provide guidance by voice under a situation where the accuracy of the low driving load on the driver is high.

In addition, the said embodiment can also be suitably changed and implemented as follows.
The guidance-related parameters only need to include the sound volume relative value and the number of mora, which are parameters related to voice, and may not include the steering angle upper limit value and noise upper limit value related to the guidance timing. In this case, the parameter that is adjusted when the response detection unit 34 detects that the driver cannot hear the guidance is the relative volume value or the number of mora, and the steering angle upper limit value and the noise upper limit value are irrelevant to the driver. The fixed value of.

-The minimum and maximum values of each parameter can be changed as appropriate.
The parameter relating to the volume may be the volume itself, not the relative volume value.
The timing adjustment unit 32 may adjust the timing so that voice guidance is performed when the driving load on the driver is low. For this reason, the timing adjustment unit 32 may adjust the timing based on, for example, the weather or the presence / absence of a parallel vehicle with respect to the surrounding situation of the vehicle. You may adjust.

The response detection unit 34 may detect the driver's response based on, for example, image information from the driver monitor camera 16 in addition to the driver's operation and speech described above.
The response detection unit 34 may detect the driver's response according to the presence or absence of the switch operation by the driver, for example, when guidance is provided for a device whose operating state changes due to the switch operation by the driver. Good.

The parameter learning unit 39 only needs to learn the initial value of each parameter, and the learning method is not limited to the method described above.
The parameter learning unit 39 selects each parameter as an adjustment parameter in the order of the relative volume value, the number of mora, the steering angle upper limit value, and the noise upper limit value, but this order can be changed as appropriate. For example, the parameter learning unit 39 may preferentially select the noise upper limit value as the adjustment parameter for a driver whose noise upper limit value is lower than a general value.

  DESCRIPTION OF SYMBOLS 11 ... Input microphone, 12 ... Turn signal, 13 ... Vehicle speed sensor, 14 ... Shift position sensor, 15 ... Steering angle sensor, 16 ... Driver monitor camera, 17 ... Input device, 18 ... Output speaker, 19 ... Display device, 30 DESCRIPTION OF SYMBOLS ... Voice guidance apparatus, 31 ... Parameter setting part, 32 ... Timing adjustment part, 33 ... Guide part, 34 ... Response detection part, 35 ... Volume adjustment part, 36 ... Speech speed adjustment part, 37 ... Steering angle upper limit value adjustment part, 38 ... Noise upper limit adjustment unit, 39 ... Parameter learning unit, 40 ... Learning data.

Claims (1)

A guidance unit that provides voice guidance to the driver of the vehicle;
A parameter setting unit for setting an initial value of a parameter relating to the guidance including the volume and speed of the voice for each driver;
A timing adjustment unit that adjusts the timing of performing the guidance by determining whether or not the driver is in a situation where it is easy to hear the voice;
A response detector for detecting a driver's response to the guidance;
A parameter adjustment unit that adjusts the parameter based on the detection result of the response detection unit;
A parameter learning unit that learns an initial value of the parameter for each driver;
When the response detection unit detects that the driver cannot hear the guidance, parameter adjustment by the parameter adjustment unit is performed so that the volume of the voice is increased or the speech speed is decreased. After performing, the guidance by the guidance unit again,
When the response detection unit detects that the driver can hear the guidance, the parameter learning unit performs learning based on a comparison result between the parameter value at that time and the initial value of the parameter. system.