JP2024067341A - Information presentation method and information presentation device for vehicle - Google Patents

Abstract

To provide an information presentation method and an information presentation device for vehicles, which prevent a driver talking with another person from missing guidance voice.SOLUTION: When a driver of a vehicle talks with another person, guidance voice which is similar to voice of the another person and presents information to the driver is generated, and the generated guidance voice is outputted from a speech position of the another person.SELECTED DRAWING: Figure 1

Description

The present invention relates to a vehicle information presentation method and information presentation device.

A navigation method is known in which acquired voice data is recorded in a voice file, the voice data is subjected to voice recognition processing to extract expressions to be used for route guidance, the extracted expressions are recorded in association with the voice file, the driving route, and the vehicle's position at the time the voice was acquired, and when traveling along the same route as the recorded driving route, if it is detected that the vehicle's position is within a predetermined range from the recorded vehicle position, the voice file corresponding to the recorded vehicle position is output (Patent Document 1).

International Publication No. 2013/051072

Even when various sounds are being produced simultaneously, humans can unconsciously select and hear the information they need and that is important to them. This is also known as the cocktail party effect.

In the above-mentioned conventional technology, when driving along the same route as the recorded driving route, the recorded audio file is output as a guidance voice that presents information to the driver within a specified range from the recorded vehicle position. If the driver is talking to someone else when this guidance voice is output, there is a problem that the driver will selectively hear only the voice uttered by the other person due to the cocktail party effect mentioned above, and miss the guidance voice.

The problem that the present invention aims to solve is to provide a vehicle information presentation method and information presentation device that can prevent a driver who is talking to someone from missing audio guidance.

The present invention solves the above problem by generating guidance voice that is similar to the voice of another person when the driver of a vehicle is talking to another person and presents information to the driver, and outputting the generated guidance voice from the position where the other person is speaking.

The present invention can prevent a driver who is engaged in a conversation with another person from missing the audio guidance.

1 is a block diagram showing an example of an information presentation system including an information presentation device according to an embodiment of the present invention. 2 is a cross-sectional view perpendicular to the height direction of the vehicle, showing an example of the arrangement of the in-vehicle camera, microphone, and speaker shown in FIG. 1 in the vehicle cabin. 2 is a flowchart showing an example of a processing procedure in the information presentation system of FIG. 1 (part 1); 10 is a flowchart showing an example of a processing procedure in the information presentation system of FIG. 1 (part 2); 10 is a flowchart showing an example of a processing procedure in the information presentation system of FIG. 1 (part 3); 10 is a flowchart showing an example of a processing procedure in the information presentation system of FIG. 1 (part 4);

The following describes an embodiment of the present invention with reference to the drawings.

[Configuration of information presentation system]
Fig. 1 is a block diagram showing an information presentation system 1 according to the present invention. The information presentation system 1 is, for example, an in-vehicle system, and as shown in Fig. 1, includes an in-vehicle camera 11, a microphone 12, a communication device 13, a navigation device 14, a speaker 15, and an information presentation device 16. The devices constituting the information presentation system 1 are connected by a Controller Area Network (CAN) or other in-vehicle LAN, and can transmit and receive information to and from each other.

The in-vehicle camera 11 is a camera installed inside the vehicle, and examples of such cameras include cameras equipped with imaging elements such as CCD and CMOS, ultrasonic cameras, and infrared cameras. The objects photographed by the in-vehicle camera 11 are objects present inside the vehicle, and are primarily occupants (specifically, the faces of the occupants). The in-vehicle camera 11 is installed in a position where it can photograph the faces of the occupants so that the occupants can be identified. The in-vehicle camera 11 may also be installed in a position where the number of occupants and their seating positions can be identified from the captured images. Note that if one in-vehicle camera 11 cannot photograph all occupants, multiple in-vehicle cameras 11 may be installed.

The microphone 12 is a device that captures sounds inside the vehicle as audio data, and examples of such microphones include a stand microphone, a close-talking microphone, and a gun microphone. The microphone 12 may be either directional or omnidirectional, and may use either a wired or wireless communication method. The sound captured by the microphone 12 is sound inside the vehicle cabin, and includes voices emitted by the occupants, and sound output from the communication device 13 and the speaker 15. The microphones 12 may be installed at appropriate positions within a range in which sound inside the vehicle cabin can be detected, and the number of microphones 12 may be determined appropriately depending on the detection range of the microphones 12 and the size of the vehicle cabin.

The communication device 13 is a device that allows the driver to communicate with people outside the vehicle, and can communicate with a communication device 13a owned by a person outside the vehicle via a network. In this embodiment, the communication device 13a is not included in the information presentation system 1. The communication device 13 is, for example, a mobile terminal such as a smartphone, and can communicate using the microphone 12 and speaker 15 mounted on the vehicle instead of the microphone and speaker provided in the communication device 13.

The navigation device 14 is a device that refers to map information (not shown) and calculates a driving route from the current position of the vehicle to a destination set by the occupant. The current position of the vehicle is acquired from a positioning system using a GPS (Global Positioning System), for example. The navigation device 14 searches for a driving route for the vehicle to reach the destination from the current position using, for example, road information and facility information from high definition map information (HD map). The driving route includes information on the road the vehicle is traveling on, the driving lane, and the vehicle's driving direction, and is displayed, for example, linearly.

As shown in FIG. 1, image data acquired by the in-vehicle camera 11, voice data acquired by the microphone 12, voice data output from the communication device 13, and guidance information for the driving route calculated by the navigation device 14 are output from each device as necessary and acquired by the information presentation device 16.

The speaker 15 is a device that converts electrical signal energy into acoustic energy and releases it into space, and is also called a loudspeaker. As shown in FIG. 1, the speaker 15 converts the guidance voice signal output from the information presentation device 16 into sound and releases (outputs) it into the vehicle cabin. The speaker 15 can be installed in an appropriate position within a range in which the sound can be transmitted to the occupants, and the number of speakers 15 is determined appropriately according to the transmission range of the speaker 15 and the size of the vehicle cabin.

An example of the arrangement of the above-mentioned in-vehicle camera 11, microphone 12, and speaker 15 in the vehicle cabin is shown in FIG. 2. FIG. 2 is a cross-sectional view perpendicular to the height direction of the vehicle V. A driver's seat A1 is provided on the right front side of the vehicle V shown in FIG. 2, a passenger seat A2 is provided on the left front side, and rear seats A3 and A4 are provided on the rear side of the vehicle V. In addition, in the vehicle V shown in FIG. 2, a driver B1 sits in the driver's seat A1, an occupant B2 sits in the passenger seat A2, an occupant B3 sits in the rear seat A3, and an occupant B4 sits in the rear seat A4. Occupants B2 to B4 are passengers who ride in the same vehicle V as the driver B1.

In vehicle V shown in FIG. 2, an in-vehicle camera 11 is installed between the driver's seat A1 and the passenger seat A2 to capture images of the driver B1 and passengers B2 to B4. Specifically, the in-vehicle camera 11 is attached to a rearview mirror installed at the top of the windshield. A microphone 12a is installed in front of the driver's seat A1 to capture the voice of driver B1, and a microphone 12b is installed in front of the passenger seat A2 to capture the voice of passenger B2. Specifically, microphone 12a is installed on the steering wheel or instrument panel of vehicle V, and microphone 12b is installed in the glove box. A microphone 12c is installed near the center of the passenger compartment on the roof of vehicle V to capture the voices of passengers B3 and B4. Furthermore, a speaker 15a is installed on the right front of the driver's seat A1 to transmit sound to the driver B1, a speaker 15b is installed on the left front of the passenger seat A2 to transmit sound to the passenger B2, a speaker 15c is installed on the right rear of the rear seat A3 to transmit sound to the passengers B3 and B4, and a speaker 15d is installed on the left rear of the rear seat A4.

Returning to FIG. 1, the information presentation device 16 is a device that controls and cooperates with each device that constitutes the information presentation system 1 to present information to vehicle occupants. The information presentation device 16 is, for example, a computer, and includes a CPU (Central Processing Unit) that is a processor, a ROM (Read Only Memory) that stores programs, and a RAM (Random Access Memory) that functions as an accessible storage device. The CPU is an operating circuit that executes the programs stored in the ROM and realizes the functions of the information presentation device 16.

The information presentation device 16 has an information presentation function that presents information to vehicle occupants. A program that realizes the information presentation function is stored in the ROM of the information presentation device 16, and the information presentation function is realized by the CPU executing the program stored in the ROM. For convenience, FIG. 1 shows the identification unit 2, acquisition unit 3, model generation unit 4, voice generation unit 5, and output unit 6 as functional blocks that realize the information presentation function. The functions of each functional block shown in FIG. 1 are explained below.

[Functions of each functional block]
The identification unit 2 has a function of identifying the number of occupants and the seating position of each occupant. The information presentation device 16 analyzes the image acquired from the in-vehicle camera 11 by the function of the identification unit 2, and judges whether or not an occupant is sitting in each seat. For example, the information presentation device 16 performs pattern matching on the image acquired from the in-vehicle camera 11, and when a pattern in which an occupant is sitting in a seat is detected, it judges that an occupant is sitting in the seat. In this case, the face of the occupant in each seat may be identified to identify the occupant.

Alternatively or in addition, the information presentation device 16 may obtain the pressure value of a pressure sensor installed in each seat and determine whether the pressure value is equal to or greater than a predetermined pressure value. The predetermined pressure value may be, for example, a pressure value measured when an adult is seated in the seat, and if the pressure value is equal to or greater than the predetermined pressure value, it is determined that an occupant is seated in the seat, and if the pressure value is less than the predetermined pressure value, it is determined that no occupant is seated in the seat.

For vehicle V shown in FIG. 2, the information presentation device 16 detects pressure values from pressure sensors installed in the driver's seat A1, passenger seat A2, and rear seats A3 and A4, and determines that the number of occupants is four because the pressure values from each pressure sensor are equal to or greater than a predetermined pressure value. The information presentation device 16 also performs pattern matching on the image acquired from the in-vehicle camera 11 to identify the faces of the driver B1 and passengers B2 to B4, and determines that the driver B1 is seated in the driver's seat A1, the passenger seat A2, the rear seat A3, and the rear seat A4.

The acquisition unit 3 has a function of determining whether the driver B1 of the vehicle V is talking to another person, and acquiring the voice of the other person when it is determined that the driver B1 is talking to another person. In this embodiment, the other person is a person with whom the driver B1 of the vehicle V can talk, and specifically, is at least one of a passenger in the vehicle V and a caller outside the vehicle who talks to the driver B1 via the communication device 13. In other words, in this embodiment, even if the function of the identification unit 2 determines that there are no occupants other than the driver B1, the driver B1 may talk to the caller.

In order to determine whether the driver B1 is talking to another person, the information presentation device 16 acquires the voice in the vehicle V from the microphone 12 using the function of the acquisition unit 3, and extracts the voice of the occupant from the acquired voice. Specifically, the information presentation device 16 performs processing to remove the voice output from the equipment of the vehicle V, the sound generated from the parts constituting the vehicle V as the vehicle V runs, and the noise outside the vehicle from the voice in the vehicle. Examples of the voice output from the equipment of the vehicle V include the radio voice output from the speaker 15 and the sound effects of the navigation device 14. Examples of the sound generated from the parts constituting the vehicle V include the sound of the tires rolling on the road surface, the operating sound of the engine, and the sound of the brakes during braking. The information presentation device 16 superimposes the voice having a waveform in the opposite phase to the waveform of these voices (sounds) on the acquired voice in the vehicle interior, and removes the voices (sounds) other than the voice of the occupant.

The information presentation device 16 also analyzes the images acquired from the in-vehicle camera 11 to detect the mouth movements of each occupant. It then compares the timing of the occupant's speech in the acquired voice with the mouth movements of the driver B1 to determine whether the driver B1 is speaking. More specifically, it determines whether the driver B1 has spoken multiple times within a certain period of time. The certain period of time can be set appropriately within a range in which it is possible to confirm that the driver B1 is speaking with another person, for example, 5 to 30 seconds. If the information presentation device 16 determines that the driver B1 has spoken multiple times within a certain period of time, it recognizes that the driver B1 is speaking with a passenger and/or a caller. On the other hand, if it determines that the driver B1 has not spoken multiple times within a certain period of time, it recognizes that the driver B1 is not speaking with a passenger and/or a caller.

For example, if driver B1 and passenger B2 are having a conversation in vehicle V shown in FIG. 2, information presentation device 16 acquires the voice in the vehicle cabin from microphones 12a to 12c, and removes the voice output from the speaker of communication device 13 and speaker 15 from the acquired voice. It also removes frictional noise and vibration noise generated when vehicle V is traveling. In this way, only the voices of driver B1 and passenger B2 are extracted from the acquired voice in the vehicle cabin. Next, it analyzes the image acquired from in-vehicle camera 11 to detect the mouth movement of driver B1. It then compares the mouth movement of driver B1 with the timing of the speech in the extracted voice, and determines that driver B1 has spoken multiple times within a certain period of time. In this way, information presentation device 16 recognizes that driver B1 is conversing with a passenger.

When it is determined that there is an occupant other than the driver B1, the information presentation device 16 may determine whether or not the driver B1 and at least one of the occupants other than the driver B1 have spoken within a certain period of time. In this case, when it is determined that the driver B1 and the occupant other than the driver B1 have spoken within a certain period of time, the information presentation device 16 recognizes that the driver B1 is talking to a passenger. In contrast, when it is determined that the driver B1 and the occupant other than the driver B1 have not spoken within a certain period of time, the information presentation device 16 recognizes that the driver B1 is not talking to a passenger. Note that when there are multiple occupants other than the driver B1, it may be determined by a similar method whether or not the occupants are talking to each other.

In addition, if it is determined that there are no occupants other than the driver B1, or if it is determined that no occupants other than the driver B1 are speaking, the information presentation device 16 may determine whether the communication device 13 is in a call with a communication device 13a external to the information presentation system 1. In this case, when it is determined that the communication device 13 is in a call with the external communication device 13a, the information presentation device 16 determines that the driver B1 is talking to the caller, and when it is determined that the communication device 13 is not in a call with the external communication device 13a, it determines that the driver B1 is not talking to either a passenger or the caller.

Furthermore, the information presentation device 16 may identify which occupant has spoken which voice based on the phase difference between the waveforms of the voices acquired by the multiple microphones 12 and the relationship with the positions at which the microphones 12 are installed. For example, when the driver B1 speaks in the vehicle V shown in FIG. 2, a voice having a certain waveform is detected by the microphone 12a, and then a voice having the same waveform is detected by the microphone 12c. Since the difference in the detection timing of the voice at the microphones 12a and 12c is proportional to the difference between the distance between the driver B1 and the microphone 12a and the distance between the driver B1 and the microphone 12c, the voice can be identified as having been spoken by the driver B1. In this case, it is possible to determine whether the driver B1 is speaking (i.e., whether the driver B1 is talking to another person) without using an image acquired from the in-vehicle camera 11.

When the information presentation device 16 determines that the driver B1 is talking to another person, it acquires the voice of the other person who is talking to the driver B1 through the function of the acquisition unit 3. When the driver B1 is talking to a passenger in the same vehicle V, the information presentation device 16 acquires the voice of the passenger, and when the driver B1 is talking to a caller outside the vehicle via the communication device 13, it acquires the voice of the caller. In contrast, when it is determined that the driver B1 is not talking to another person, the information presentation device 16 repeats the determination of whether the driver B1 is talking to another person. Alternatively, the information presentation device 16 goes into a standby state. Hereinafter, the voice of the passenger is also referred to as the first voice, and the voice of the caller is also referred to as the second voice.

The first voice of the passenger is obtained by extracting the voice of the passenger from the voice in the vehicle cabin in the same manner as described above. For example, when the driver B1 and the passenger B2 are talking in the vehicle V shown in FIG. 2, the voice in the vehicle cabin is obtained from the microphones 12a to 12c, and the voice output from the speaker 15 and the like are removed from the obtained voice, and only the voices of the driver B1 and the passenger B2 are extracted. Next, the mouth movement of the passenger B2 is detected from the image obtained from the in-vehicle camera 11, and only the part where the mouth of the passenger B2 is open (i.e., the passenger B2 is speaking) is further extracted from the voice of only the driver B1 and the passenger B2. In contrast, the second voice of the caller is obtained by obtaining the voice of the caller output from the speaker of the communication device 13 or the speaker 15 of the vehicle V.

The model generation unit 4 has a function of generating a voice model for generating guidance voice. A voice model is a model in which voice features are set, and is, for example, software that reads input text and generates an artificial voice. Features are numerical values that indicate the characteristics of a voice, and specific examples include pronunciation speed, voice volume, voice waveforms of vowels and consonants, frequency of word use, types of dialects and their frequency of use.

These features are obtained by morphological analysis of the spoken content, frequency analysis of the voice waveform, and other analyses. In particular, the information presentation device 16 performs frequency analysis on the acquired voice, and calculates the frequency components and their amplitudes obtained by the frequency analysis as features. Frequency analysis refers to determining the frequency components and their amplitudes contained in a waveform by breaking down the waveform into multiple simple sine waves and cosine waves. The information presentation device 16 performs the above-mentioned various analyses within a predetermined range of the acquired voice waveform. The predetermined range is, for example, the range from when the occupant who is talking to the driver B1 starts speaking to when the driver B1 or another occupant starts speaking. The predetermined range may also be the range from when the occupant who is talking to the driver B1 starts speaking to when he finishes speaking a sentence.

The information presentation device 16 uses the functions of the model generation unit 4 to perform frequency analysis and other analyses on the voice, calculates the features of the voice, and generates a voice model that can generate a guidance voice similar to the acquired voice. The information presentation device 16 may generate a new voice model using machine learning (e.g., deep learning) based on the calculated features, or may generate a voice model by modifying a pre-registered voice model using the calculated features. Furthermore, the information presentation device 16 may select a voice model with the most similar features from multiple pre-registered voice models based on the features. By selecting one model from the pre-registered voice models, the time-consuming process of generating a voice model can be omitted.

The feature quantities being most similar means that the difference between the feature quantities is the smallest (i.e., they are the most similar). For example, this means that the difference between feature quantities (particularly frequency components and their amplitudes) is the smallest. As an example, when comparing frequency components and their amplitudes in order from the smallest frequency components, the feature quantity with the smallest sum of the differences when the differences in frequency are added is the most similar feature quantity. As another example, when comparing amplitude differences in order from the smallest frequency components, the feature quantity with the smallest sum of the differences when the differences in amplitude are added is the most similar feature quantity.

For example, in the case where the driver B1 and the occupant B2 are having a conversation in the vehicle V shown in FIG. 2, when the voice of the occupant B2 is acquired, the information presentation device 16 performs a frequency analysis on the voice of the occupant B2 and calculates the frequency components and their amplitudes contained in the voice of the occupant B2. Then, based on the frequency components and the amplitudes, the default voice model of the navigation device 14 is changed, and a voice model is generated for generating a guidance voice similar to the voice of the occupant B2.

The voice generating unit 5 has a function of generating guidance voice that presents information to the driver B1. The information presented to the driver B1 is, for example, information output from the navigation device 14, and in particular, information for the vehicle V to travel along a set travel route. Specifically, this includes the travel state of the vehicle V, whether or not to turn right or left at an intersection, whether or not there is a dedicated lane for right or left turns, whether or not to change lanes, and the location of obstacles that need to be avoided. In addition, the information presented by the guidance voice includes information other than information for traveling along the travel route, such as tourist information near the destination, information about nearby facilities, current events such as news, and information about malfunctions of the vehicle V. This information is collectively referred to as guidance information.

In order to convey the above-mentioned guidance information to the driver B1, the information presentation device 16 generates a guidance voice using a voice model through the function of the voice generation unit 5. Specifically, at least one of a guidance voice similar to the passenger's first voice and a guidance voice similar to the caller's second voice is generated. The content of the generated guidance voice is based on the guidance information acquired from the navigation device 14, as shown in FIG. 1. Hereinafter, the guidance voice similar to the passenger's first voice is also referred to as the first guidance voice, and the guidance voice similar to the caller's second voice is also referred to as the second guidance voice.

The guidance voice generated by the information presentation device 16 is similar to the voice of another person who is conversing with the driver B1. The guidance voice being similar to the voice of another person includes the guidance voice being identical to the voice of the other person, and means that when a guidance voice is output to the driver B1 who is conversing with another person, the cocktail party effect is suppressed, and it is possible to prevent the driver B1 from missing the guidance voice. In other words, if the voice is similar to the voice of the passenger who is conversing, the driver B1 can unconsciously select and hear the guidance voice even during a conversation.

As an example, frequency analysis is performed on the waveform of the other person's voice and the generated guidance voice within the above-mentioned specified range to calculate the frequency components and their amplitudes. Next, the frequency components contained in the other person's voice and the frequency components contained in the guidance voice are associated with each other in ascending order of value. Next, the amplitude of the frequency components of the other person's voice is compared with the amplitude of the frequency components of the guidance voice corresponding to the frequency components of the other person's voice. Then, for each corresponding frequency component, it is determined whether the amplitude of the guidance voice is within a range of ±10% of the amplitude of the other person's voice. If it is determined that the amplitude of the guidance voice is within a range of ±10% of the amplitude of the other person's voice, it is determined that the guidance voice is similar to the other person's voice. On the other hand, if it is determined that the amplitude of the guidance voice is within a range of less than -10% or more than +10% of the amplitude of the other person's voice, it is determined that the guidance voice is not similar to the other person's voice. Note that the frequency analysis of the other person's voice is performed within the above-mentioned specified range, and the frequency analysis of the guidance voice is performed on the entire generated guidance voice.

As another example, the frequency components contained in the voice of another person and the frequency components contained in the guidance voice are matched in ascending order of value. Then, it is determined whether the maximum value of the difference between the two corresponding frequency components is 5 to 50 [Hz] or less. If it is determined that the maximum value of the difference between the two corresponding frequency components is 5 to 50 [Hz] or less, it is determined that the guidance voice is similar to the voice of another person. In contrast, if it is determined that the maximum value of the difference between the two corresponding frequency components exceeds 5 to 50 [Hz], it is determined that the guidance voice is not similar to the voice of another person. Note that the range in which frequency analysis is performed on the voice of another person and the guidance voice is the same as in the above example.

For example, in the case where the driver B1 and the passenger B2 are having a conversation in the vehicle V shown in FIG. 2, when the voice of the passenger B2 is acquired and a voice model for generating a guidance voice similar to the voice of the passenger B2 is generated, the information presentation device 16 acquires guidance information from the navigation device 14. If the guidance information is information that it is necessary to turn right at the intersection ahead, the information presentation device 16 uses the voice model based on the acquired guidance information to generate a guidance voice such as "Turn right at the next intersection." The guidance voice is similar to the voice of the passenger B2, and as an example, for each frequency component included in the voice of the passenger B2, the frequency component with the closest value is selected from the frequency components included in the guidance voice, and the two frequency components are matched, and the maximum value of the difference between the two corresponding frequency components is 5 to 50 [Hz] or less.

The output unit 6 has a function of outputting the guidance voice generated by the function of the voice generation unit 5 from the speaking position of the other person who is talking to the driver B1. The speaking position of the other person is the position where the voice of the other person is emitted (or output), for example, the position of the head of the other person, and particularly the position of the mouth. If the other person is a passenger, the speaking position is, for example, the seating position of the passenger, more specifically, the position of the head of the passenger. When the driver B1 is talking to the passenger, the information presentation device 16 outputs the first guidance voice from the passenger to the driver by the function of the output unit 6. On the other hand, when the other person is the caller, the speaking position is a speaker that outputs the second voice, for example, the speaker of the communication device 13 or the speaker 15 of the vehicle V. When the driver B1 is talking to the caller, the information presentation device 16 outputs the second guidance voice from the speaker that outputs the second voice to the driver by the function of the output unit 6.

For example, in the case where the driver B1 and the passenger B2 are having a conversation in the vehicle V shown in FIG. 2, when a guidance voice similar to the voice of the passenger B2 is generated, the information presentation device 16 outputs the guidance voice to the driver B1 (or the driver's seat A1) from the speaker 15b installed on the left front side of the passenger seat A2. This allows the driver B1 to recognize that the guidance voice is being spoken by the passenger B2, and can unconsciously select and hear the guidance voice even when the driver B1 is having a conversation with the passenger B2.

The information presentation device 16 may use the function of the output unit 6 to output a guidance voice from the vicinity of the seating position of the passenger who is conversing with the driver B1. For example, out of the multiple speakers 15, the guidance voice may be output from the speaker closest to the seating position of the passenger. Furthermore, the information presentation device 16 may use the function of the output unit 6 to control the speaker 15 in the vehicle so that the driver B1 recognizes that the passenger has spoken. Specifically, the information presentation device 16 controls the volume of the voice output from the speaker 15, the phase of the voice waveform, the output timing, etc. Furthermore, the position at which the speaker 15 is installed in the vehicle may be set so that the driver B1 recognizes that the passenger has spoken.

For example, in the vehicle V shown in FIG. 2, when the driver B1 and passenger B2 are having a conversation, a guidance voice similar to the voice of passenger B2 is output from speakers 15b and 15d to the driver B1. In this case, the phase of the waveform of the guidance voice output from one of the speakers is controlled so that the guidance voice output from speaker 15b and the guidance voice output from speaker 15d do not cancel each other out.

The information presentation device 16 may use the function of the output unit 6 to determine whether the ratio of speech time between the driver B1 and another person who is conversing with the driver B1 in a predetermined time period is equal to or less than a predetermined value. The predetermined time period can be set appropriately within a range in which the timing for interrupting the conversation between the driver B1 and another person with a guidance voice can be grasped, for example, 15 to 60 seconds. The predetermined value can be set appropriately within a range in which the driver B1 is prevented from missing the guidance voice when the guidance voice is interrupted in the conversation between the driver B1 and another person, for example, 0 to 20%. When the information presentation device 16 determines that the ratio of speech time between the driver B1 and another person in a predetermined time period is equal to or less than a predetermined value, it outputs the guidance voice at the timing when it is determined that the ratio is equal to or less than the predetermined value.

In contrast, if it is determined that the ratio of speech time of driver B1 and others to a predetermined time exceeds a predetermined value, the guidance voice is not output. Alternatively, if the information presentation device 16 determines that the ratio of speech time of driver B1 and others to a predetermined time exceeds a predetermined value, it may determine whether the speaker is driver B1 or not, and output the guidance voice if it is determined that the speaker is driver B1. In contrast, if it is determined that the speaker is not driver B1, the guidance voice is not output. Alternatively, the information presentation device 16 may output the guidance voice immediately at the timing when the guidance voice is generated. In other words, the guidance voice may be output at the same time as the generation of the guidance voice is completed.

For example, in the case where driver B1 and passenger B2 are having a conversation in vehicle V shown in FIG. 2, if the voice in the vehicle cabin is acquired for 30 seconds and it is detected that the speech time of driver B1 and passenger B2 is 3 seconds, the ratio of speech time to the specified time is 10%, so the information presentation device 16 outputs a guidance voice at a timing calculated based on this ratio. Note that a state in which the ratio of speech time to the specified time is equal to or less than a specified value is also referred to as a state in which the conversation between driver B1 and another person has been interrupted.

The information presentation device 16 may obtain guidance information from the navigation device 14 using the function of the output unit 6, and may determine whether the guidance voice based on the guidance information is a voice requesting an operation from the driver B1. The voice requesting an operation is, for example, a voice instructing the driver B1 to perform a driving operation, such as a guidance voice instructing the driver B1 to turn right or left at an intersection, a guidance voice instructing the driver to enter a right or left turn exclusive lane, and a guidance voice notifying the driver of the presence of an approaching vehicle. In contrast, a voice that does not request an operation is a voice that simply presents information to the driver B1, such as a voice informing the driver B1 of the existence of a parking lot nearby and a voice conveying news. If it is determined that the guidance voice is a voice requesting an operation, the information presentation device 16 outputs the voice requesting an operation in priority over other voices that do not request an operation. In contrast, if it is determined that the guidance voice is not a voice requesting an operation, the information presentation device 16 outputs the other voices that request an operation in priority, and then outputs a voice that does not request an operation.

When outputting the guidance voice, the information presentation device 16 may use the function of the acquisition unit 3 to determine whether or not another person who is conversing with the driver B1 is speaking. For example, the information presentation device 16 analyzes an image acquired from the in-vehicle camera 11 and detects the mouth movement of the occupant who is conversing with the driver B1. Then, based on the mouth movement of the occupant (e.g., the way the mouth is open), it determines whether or not the occupant is speaking. The information presentation device 16 also acquires the second voice of the caller from the communication device 13 and determines whether or not the caller is speaking.

If it is determined that the other person who is conversing with the driver B1 is speaking, the output unit 6 functions to output a different voice having a waveform in the opposite phase to the waveform of the voice of the other person. This cancels out the voice of the other person, preventing the driver B1 from missing the guidance voice. In contrast, if it is determined that the other person who is conversing with the driver B1 is not speaking, a guidance voice similar to the voice of the other person is output. Note that if the other person is a passenger, the voice having the opposite phase waveform is output, for example, from the speaker 15 of the vehicle V, and if the other person is the caller, the voice having the opposite phase waveform is output from the speaker that outputs the second voice.

For example, in the case where driver B1 and occupant B2 are having a conversation in vehicle V shown in FIG. 2, information display device 16 detects the movement of occupant B2's mouth and determines whether occupant B2 is speaking from the movement of occupant B2's mouth. In this case, since it is determined that occupant B2 is speaking, information display device 16 outputs a different voice having a waveform in the opposite phase to the waveform of the voice spoken by the occupant from speaker 15a or speaker 15b.

When the information presentation device 16 generates a voice model of another person who has conversed with the driver B1 using the function of the model generation unit 4, the information presentation device 16 may register the generated voice model in association with the other person. For example, the voice model is registered in advance in a recording medium such as a ROM of the information presentation device 16 or in a server external to the information presentation system 1. Then, when the information presentation device 16 acquires the voice of another person who is conversing with the driver B1 using the function of the acquisition unit 3, it determines whether or not the other person identified by the function of the identification unit 2 has conversed with the driver B1 in the past. For example, if an image of the face of the other person has been registered in advance in association with the voice model, it acquires an image from the in-vehicle camera 11 and recognizes the face of the other person from the acquired image. Then, it determines whether or not there is an image of a face that matches the face of the other person who is riding in the vehicle V among the pre-registered images.

If it is determined that there is a facial image in the pre-registered images that matches the face of the other person in the vehicle V, the information presentation device 16 recognizes that the other person in the vehicle V has previously spoken with the driver B1, and outputs a pre-registered voice model corresponding to that other person to the voice generation unit 5. The output voice model is then used to generate a guidance voice. On the other hand, if it is determined that there is no facial image in the pre-registered images that matches the face of the other person in the vehicle V, it recognizes that the other person in the vehicle V has not previously spoken with the driver B1, and generates a new voice model.

For example, when driver B1 and occupant B2 are having a conversation in vehicle V shown in FIG. 2, information presentation device 16 identifies the face of occupant B2 from the image from in-vehicle camera 11, and determines whether an image of the same face as the identified face of occupant B2 exists in pre-registered images. If it is determined that an image of the same face as the identified face of occupant B2 exists in pre-registered images, a guidance voice is generated using a registered voice model corresponding to occupant B2. On the other hand, if it is determined that an image of the same face as the identified face of occupant B2 does not exist in pre-registered images, a new voice model is generated using the voice features of occupant B2.

In addition, if there are multiple other people conversing with the driver B1, the information presentation device 16 may use the function of the model generation unit 4 to select the other person who has spoken the longest, and use the function of the voice generation unit 5 to generate a guidance voice similar to the voice of the other person who has spoken the longest. The speech time may be, for example, the speech time from when it is determined that the driver B1 is conversing with another person to when the generation of the voice model is completed. Alternatively, it may be the speech time from when it is determined that the driver B1 is conversing with another person to when the above-mentioned predetermined time (for example, 15 to 60 seconds) has passed. The information presentation device 16 uses the function of the output unit 6 to output the guidance voice from the speaking position of the other person who has spoken the longest.

Similarly, if there are multiple other people conversing with the driver B1, the information presentation device 16 may generate a guidance voice for each other person using the function of the voice generation unit 5, and use the function of the output unit 6 to select from the generated guidance voices the guidance voice that is most similar to the voice of the other person, and output the selected guidance voice. "Most similar to the voice of the other person" means that the difference between the frequency components and their amplitudes contained in the voice of the other person and the frequency components and their amplitudes contained in the generated guidance voice is the smallest. Note that the frequency analysis to calculate the frequency components and their amplitudes is performed on the entire generated guidance voice.

Similarly, if there are multiple other people conversing with the driver B1, the information presentation device 16 generates a guidance voice for each other person using the function of the voice generation unit 5, and determines whether or not there is a non-speaking other person among the other people using the function of the acquisition unit 3. The information presentation device 16, for example, analyzes images acquired from the in-vehicle camera 11 to detect the mouth movement of the occupant conversing with the driver B1. Then, based on the mouth movement of the occupant (e.g., the opening of the mouth), determines whether or not the occupant is speaking. If the information presentation device 16 determines that there is a non-speaking other person, it outputs a guidance voice for the non-speaking other person. On the other hand, if it determines that there is no non-speaking other person, it stops outputting the guidance voice once, and repeats the determination until it determines that there is a non-speaking other person.

[Processing in Information Presentation System]
The procedure for processing information by the information presentation device 16 will be described with reference to Figures 3 to 6. Figures 3 to 6 are examples of flowcharts showing information processing executed in the information presentation system 1 of this embodiment. The processing described below is executed at predetermined time intervals by the processor (CPU) of the information presentation device 16.

The flowcharts shown in Figures 3 to 6 are based on the premise that the vehicle V is being driven manually by the driver B1. Manual driving refers to controlling the vehicle's driving by the driver's operation, rather than by on-board equipment.

First, in step S1 of FIG. 3, the identification unit 2 functions to identify the number of occupants in the vehicle V and the seating positions of each occupant from the image captured by the in-vehicle camera 11. In the following step S2, the acquisition unit 3 functions to acquire voices within the vehicle cabin from the microphone 12, and in step S3, the mouth movements of the occupants (particularly the driver B1) are detected from the image captured by the in-vehicle camera 11. In step S4, the acquisition unit 3 functions to determine whether the driver B1 is talking to another person. If it is determined that the driver B1 is not talking to another person, the process proceeds to step S2. On the other hand, if it is determined that the driver B1 is talking to another person, the process proceeds to step S5.

In step S5, it is determined whether the other person with whom the driver B1 is talking is a passenger in the vehicle V. If it is determined that the other person is not a passenger in the vehicle V, the process proceeds to step S21 in FIG. 4. On the other hand, if it is determined that the other person is a passenger in the vehicle V, the process proceeds to step S6. FIG. 4 will be described later. In step S6, it is determined whether there are multiple passengers. If it is determined that there are multiple passengers, the process proceeds to step S41 in FIG. 5 or step S61 in FIG. 6. On the other hand, if it is determined that there are not multiple passengers, the process proceeds to step S7.

In step S7, the function of the acquisition unit 3 is to remove the voice output from the vehicle equipment from the voice within the vehicle cabin, and in the subsequent step S8, the first voice is extracted from the voice within the vehicle cabin. In step S9, frequency analysis is performed on the first voice, and in the subsequent step S10, the feature quantities of the first voice are acquired. In step S11, the function of the model generation unit 4 is to generate a voice model using the feature quantities, and in the subsequent step S12, the function of the voice generation unit 5 is to generate the first guidance voice using the generated voice model.

In step S13, the function of the output unit 6 is used to determine whether the ratio of speech time taken by the driver B1 and other persons in a given time period is equal to or less than a given value. If it is determined that the ratio is equal to or less than the given value, the process proceeds to step S14, where a first guidance voice is output from the passenger to the driver. Then, execution of this routine ends. On the other hand, if it is determined that the ratio exceeds the given value, step S13 is repeated.

Next, a case where it is determined in step S5 of FIG. 3 that the other person is not a passenger in vehicle V will be described with reference to FIG. 4.

First, in step S21 of FIG. 4, the function of the acquisition unit 3 is used to acquire the second voice from the communication device 13, and in the subsequent step S22, frequency analysis is performed on the second voice, and in the subsequent step S23, the feature quantities of the second voice are acquired. In step S24, the function of the model generation unit 4 is used to determine whether the caller has previously spoken with the driver B1. If it is determined that the caller has previously spoken with the driver B1, the process proceeds to step S25, where the second guidance voice is generated using a pre-registered voice model. On the other hand, if it is determined that the caller has not previously spoken with the driver B1, the process proceeds to step S26, where a voice model is generated using the feature quantities, and in the subsequent step S27, the function of the voice generation unit 5 is used to generate the second guidance voice using the generated voice model.

In step S28, the function of the output unit 6 is used to determine whether the caller is speaking. If it is determined that the caller is speaking, the process proceeds to step S29, where a different voice having a waveform in the opposite phase to the waveform of the second voice is output. Then, the process proceeds to step S30. On the other hand, if it is determined that the caller is not speaking, the process proceeds to step S30, where a second guidance voice is output to the driver from the speaker that outputs the second voice. Then, execution of this routine ends.

Next, the case where it is determined in step S6 of FIG. 3 that there are multiple passengers will be described with reference to FIG. 5.

First, in step S41 of FIG. 5, the first voice of each passenger is extracted from the voice in the vehicle cabin. In step S42, frequency analysis is performed on the first voice of each passenger, and in the subsequent step S43, the features of the first voice of each passenger are obtained. In step S44, the function of the model generation unit 4 is used to select a voice model having features most similar to the features of the first voice of each passenger, and in the subsequent step S45, the function of the voice generation unit 5 is used to generate a first guidance voice using the selected voice model. In the subsequent step S46, it is determined whether or not there is a passenger who is not speaking.

If it is determined that there is a passenger who is not speaking, the process proceeds to step S47, where the output unit 6 uses its function to output a first guidance voice for the passenger who is not speaking, and in step S48, the speaker in the vehicle is controlled so that it is recognized that the passenger has spoken. Then, execution of this routine is terminated. On the other hand, if it is determined that there is no passenger who is not speaking, the process proceeds to step S49, where the first guidance voice that is most similar to the first voice is selected, and in the following step S50, the selected first guidance voice is output to the driver B1 from the speaker 15 located closest to the passenger's seating position. Then, execution of this routine is terminated.

Next, in the case where it is determined in step S6 of FIG. 3 that there are multiple passengers, the process that differs from the process shown in FIG. 5 will be described with reference to FIG. 6.

First, in step S61 of FIG. 6, the passenger with the longest speaking time is selected using the function of the model generation unit 4, then in step S62, the first voice of the passenger with the longest speaking time is extracted, then in step S63, frequency analysis is performed on the extracted first voice, and then in step S64, the feature quantities of the first voice of the passenger with the longest speaking time are acquired. In step S65, the voice model to which the acquired feature quantities are most similar is selected, and in step S66, the first guidance voice is generated using the selected voice model using the function of the voice generation unit 5.

In step S67, the output unit 6 determines whether the guidance voice is a voice requesting an operation from the driver. If it is determined that the guidance voice is a voice requesting an operation from the driver, the process proceeds to step S68, where the first guidance voice is preferentially output from the passenger who has been speaking the longest to the driver B1. On the other hand, if it is determined that the guidance voice is not a voice requesting an operation from the driver, the process proceeds to step S69, where another guidance voice requesting an operation from the driver is preferentially output.

In step S70, it is determined whether or not another voice guidance requesting an operation has been output to the driver B1. If it is determined that another voice guidance requesting an operation has been output to the driver B1, the process proceeds to step S71, in which a first voice guidance request is output to the driver B1 from the passenger who has spoken the longest. Then, execution of this routine ends. On the other hand, if it is determined that another voice guidance requesting an operation has not been output, the process proceeds to step S69. Note that the information presentation device 16 according to this embodiment can be applied not only to cases in which the driver B1 is having a conversation with another person, but also to cases in which other people are having a conversation with each other.

[Embodiments of the invention]
As described above, according to the present embodiment, there is provided a method for presenting information to a vehicle V, which is executed by a processor, in which, when a driver B1 of the vehicle V is talking to another person, the processor acquires the voice of the other person, generates a guidance voice similar to the voice and presents information to the driver B1, and outputs the guidance voice from a position where the other person is speaking. This makes it possible to prevent the driver B1, who is talking to another person, from missing the guidance voice without impeding the driving operation of the driver B1.

According to the information presentation method of this embodiment, when the other person is a passenger in the vehicle V, the processor acquires a first voice of the passenger, generates a first guidance voice similar to the first voice and presents information for the vehicle V to travel along a travel route to the driver B1, and outputs the first guidance voice from the passenger to the driver B1. This makes it possible to prevent the driver B1 from missing the guidance voice while talking to the passenger, without interfering with the driving operation of the driver B1.

Furthermore, according to the information presentation method of this embodiment, when the other person is a caller outside the vehicle who is talking to the driver B1 via the communication device 13, the processor acquires the second voice of the caller, generates a second guidance voice similar to the second voice and presents information for the vehicle V to travel along a travel route to the driver B1, and outputs the second guidance voice to the driver B1 from a speaker that outputs the second voice. This makes it possible to prevent the driver B1, who is talking to the caller, from missing the guidance voice without interfering with the driving operation of the driver B1.

In addition, according to the information presentation method of this embodiment, the processor performs frequency analysis on the voice to calculate the features of the voice, selects the voice model with the most similar features from a plurality of pre-registered voice models based on the features, and generates the guidance voice using the selected voice model. This makes it possible to omit the step of generating a voice model.

In addition, according to the information presentation method of this embodiment, when there are multiple other people, the processor selects the other person who has spoken the longest, generates the guidance voice similar to the voice of the other person who has spoken the longest, and outputs the guidance voice from the speaking position of the other person who has spoken the longest. This makes it possible to use the voice of the person who is at the center of the conversation, and to output a guidance voice that is most easily noticeable to the driver B1.

In addition, according to the information presentation method of this embodiment, when there are multiple other people, the processor generates the guidance voice for each of the other people, selects the guidance voice that is most similar to the voice from the guidance voices, and outputs the guidance voice that is most similar to the voice. This makes it possible to suppress the discomfort that the driver B1 feels from the guidance voice.

In addition, according to the information presentation method of this embodiment, the processor determines whether the ratio of speech time of the driver B1 and the other person in a given time period is equal to or less than a given value, and outputs the guidance voice at the timing when it is determined that the ratio is equal to or less than the given value. This makes it possible to further prevent the driver B1 who is currently talking from missing the guidance voice.

In addition, according to the information presentation method of this embodiment, the processor determines whether the guidance voice is a voice requesting the driver B1 to perform an operation, and if it determines that the guidance voice is a voice requesting the operation, outputs the voice requesting the operation in priority to the voice that does not request the operation. This allows information of high importance to be transmitted with priority, and also prevents conversation from being interrupted.

In addition, according to the information presentation method of this embodiment, when the processor outputs the guidance voice, it determines whether the other person is speaking, and when it determines that the other person is speaking, it outputs a different voice having a waveform in the opposite phase to the waveform of the voice. This makes it possible to further prevent the driver B1 who is talking from missing the guidance voice.

In addition, according to the information presentation method of this embodiment, the processor pre-registers a voice model of the other person who has conversed with the driver B1 in association with the other person, determines whether the other person has conversed with the driver B1 in the past, and if it determines that the other person has conversed with the driver B1 in the past, generates the guidance voice using the pre-registered voice model corresponding to the other person. This makes it possible to omit the process of generating a voice model.

In addition, according to the information presentation method of this embodiment, when there are multiple other people, the processor generates the guidance voice for each of the other people, determines whether or not there is a other person who is not speaking among the other people, and when it determines that there is a other person who is not speaking, outputs the guidance voice for the other person who is not speaking. This makes it possible to further prevent the driver B1 who is talking from missing the guidance voice.

Furthermore, according to the information presentation method of this embodiment, the processor identifies the number of passengers and the seating positions of each passenger. This makes it possible to confirm that multiple people are riding in the vehicle V.

Furthermore, according to the information presentation method of this embodiment, the processor acquires the voice in the vehicle cabin of the vehicle V and extracts the first voice from the voice in the vehicle cabin. This reduces noise that affects the generation of the voice model, and improves the similarity of the guidance voice.

In addition, according to the information presentation method of this embodiment, when extracting the first voice from the voice within the vehicle cabin, the processor performs a process of removing the voice output from the device of the vehicle V from the voice within the vehicle cabin. This makes it possible to more accurately reduce noise that affects the generation of the voice model.

In addition, according to the information presentation method of this embodiment, the processor controls the speaker 15 in the vehicle so that the driver B1 recognizes that the passenger has spoken. This allows multiple speakers 15 to be linked together, making it possible to further reduce the likelihood that the driver B1, who is having a conversation, will miss the voice guidance.

In addition, according to the information presentation method of this embodiment, the processor outputs the guidance voice from the speaker 15 located closest to the passenger's seating position. This allows the position from which the guidance voice is output to be closer to the passenger, making it possible to further reduce the likelihood that the driver B1, who is currently having a conversation, will miss the guidance voice.

In addition, according to this embodiment, an information presentation device 16 for a vehicle V is provided, which includes an acquisition unit 3 that acquires the voice of another person when the driver B1 of the vehicle V is talking to another person, a voice generation unit 5 that generates a guidance voice that presents information to the driver B1 and is similar to the voice, and an output unit 6 that outputs the guidance voice from the speaking position of the other person. This makes it possible to prevent the driver B1, who is talking to another person, from missing the guidance voice without interfering with the driving operation of the driver B1.

Reference Signs List 1...Information presentation system 11...In-vehicle camera 12, 12a, 12b, 12c...Microphone 13, 13a...Communication device 14...Navigation device 15, 15a, 15b, 15c, 15d...Speaker 16...Information presentation device 2...Identification unit 3...Acquisition unit 4...Model generation unit 5...Speech generation unit 6...Output unit A1...Driver's seat A2...Passenger seat A3, A4...Rear seat B1...Driver B2, B3, B4...Occupant V...Vehicle

Claims (17)

A method for displaying information about a vehicle, the method comprising:
The processor,
When the driver of the vehicle is talking to another person, the voice of the other person is acquired;
generating a guidance voice similar to the voice for presenting information to the driver;
The vehicle information presentation method includes outputting the guidance voice from the speaking position of the other person. The processor,
When the other person is a passenger in the vehicle, a first voice of the passenger is acquired;
generating a first guidance voice similar to the first voice and presenting information to the driver for the vehicle to travel along a travel route;
The vehicle information presentation method according to claim 1 , wherein the first guidance voice is output from the passenger to the driver. The processor,
When the other person is a caller outside the vehicle who is talking to the driver via a communication device, a second voice of the caller is acquired;
generating a second guidance voice similar to the second voice and presenting information to the driver for the vehicle to travel along a travel route;
The vehicle information presentation method according to claim 1 , further comprising: outputting the second guidance voice to the driver from a speaker that outputs the second voice. The processor,
performing a frequency analysis on the voice to calculate a feature amount of the voice;
Selecting a voice model having the most similar feature from a plurality of pre-registered voice models based on the feature;
The method for presenting information to a vehicle according to claim 1 , further comprising generating the guidance voice using the selected voice model. The processor,
If there are multiple other people, select the other person who has spoken the longest,
generating the guidance voice similar to the voice of the other person having the longest speaking time;
The method for presenting information to a vehicle according to claim 1 , further comprising: outputting the guidance voice from the speaking position of the other person whose speech duration is the longest. The processor,
When there are a plurality of other people, the guidance voice is generated for each of the other people;
Selecting the guidance voice that is most similar to the voice from among the guidance voices;
The method for presenting information to a vehicle according to claim 1 or 4, further comprising: outputting the guidance voice that is most similar to the voice. The processor,
determining whether a ratio of speech time of the driver and the other person in a predetermined time period is equal to or less than a predetermined value;
The information presentation method for a vehicle according to claim 1 , further comprising: outputting the guidance voice at a timing when it is determined that the ratio is equal to or smaller than the predetermined value. The processor,
determining whether the guidance voice is a voice requesting an operation from the driver;
The vehicle information presentation method according to claim 1 or 4, wherein, when it is determined that the guidance voice is a voice requesting the operation, the voice requesting the operation is output in priority to a voice not requesting the operation. The processor,
When outputting the guidance voice, it is determined whether or not the other person is speaking;
5. The method for presenting information to a vehicle according to claim 1, further comprising the step of outputting a different voice having a waveform inversely phase to a waveform of the voice when it is determined that the other person is speaking. The processor,
A voice model of the other person who has had a conversation with the driver is registered in advance in association with the other person;
determining whether the other person has had a previous conversation with the driver;
5. The vehicle information presentation method according to claim 1, wherein if it is determined that the other person has had a conversation with the driver in the past, the guidance voice is generated using a pre-registered voice model corresponding to the other person. The processor,
When there are a plurality of other people, the guidance voice is generated for each of the other people;
determining whether or not there is a non-speaking other person among the other people;
The method for presenting information to a vehicle according to claim 1 or 4, further comprising the step of outputting the guidance voice of the other person who is not speaking when it is determined that the other person who is not speaking is present. The processor,
The vehicle information presentation method according to claim 2 , further comprising identifying the number of passengers and the seating positions of each passenger. The processor,
Acquire voice in the cabin of the vehicle;
The method for presenting information to a vehicle according to claim 2 or 12, further comprising extracting the first voice from a voice within the vehicle cabin. The processor,
The vehicle information presentation method according to claim 13 , further comprising the step of: removing a sound output from a device of the vehicle from the sound within the vehicle cabin when the first sound is extracted from the sound within the vehicle cabin. The processor,
The information presentation method for a vehicle according to claim 2 or 12, further comprising controlling a speaker in the vehicle so as to make the driver recognize that the passenger has spoken. The processor,
The information presentation method for a vehicle according to claim 2 or 12, wherein the guidance voice is output from a speaker located closest to a seating position of the passenger. an acquisition unit that acquires a voice of another person when the driver of the vehicle is talking to the other person;
a voice generating unit that generates a guidance voice similar to the voice and presents information to the driver;
and an output unit that outputs the guidance voice from the speaking position of the other person.

Images