JPH1165587A - Vehicular voice input device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicular voice input device, even when plural crewmen a board a vehicle of which device allows the respective crewmen to properly operate on-vehicle equipment by voice. SOLUTION: A voice signal adjusting section 2 specifies whether the voicing person is a driver or a crewman in the assistant driver's seat, based on the voice signals SIN1, SIN2 inputted through microphones 1A, 1B to output uttering person information SPK and also the voice signal on the side of the uttering person as an output signal SOUT. A voice-recognizing section 3 recognizes the content of the signal SOUT and specifies an equipment to be operated and the content of its operation. When the operation by the specified uttering person is permitted, a control signal CTL corresponding to the operation is outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle voice input device for recognizing the content of voice uttered by a vehicle occupant and inputting a control signal to on-vehicle equipment.

[0002]

2. Description of the Related Art When an occupant of a vehicle makes a call using a communication device, a voice input by a occupant is converted to an electric signal (e.g. A microphone that converts the sound signal from the microphone into a sound signal is provided on the shoulder belt (a belt that forms part of the seat belt and restrains the chest of the occupant) at a position corresponding to the chest of the occupant. 2. Description of the Related Art A voice input device adapted to input to a communication device is conventionally known.

[0003]

However, when the occupant of the vehicle recognizes the content of the voice uttered and operates the vehicle-mounted device by voice, if a microphone is provided in the driver's seat as in the above-described conventional device, Although the driver of the vehicle can operate the in-vehicle device by voice, it is difficult for an occupant on the passenger side to operate the in-vehicle device by voice, for example.
In addition, simply installing a microphone on the passenger seat side does not make it clear whether the operation is performed by the driver or by the occupant in the passenger seat, and thus an inappropriate operation may be performed.

The present invention has been made in view of this point, and is intended for a vehicle in which each occupant can appropriately operate a vehicle-mounted device by voice even when a plurality of occupants are in the vehicle. It is an object to provide a voice input device.

[0005]

According to the first aspect of the present invention, there is provided a vehicle for recognizing the content of a voice uttered by a vehicle occupant and inputting a control signal to an on-vehicle device. A plurality of microphones arranged corresponding to a plurality of seats of the vehicle, and a speaker specifying a seat on which the speaker is seated based on audio signals input from the plurality of microphones. Identification means,
Voice recognition means for recognizing the content of the voice signal input from the microphone corresponding to the seat specified by the speaker specifying means, wherein the voice recognition means, from the content of the recognized voice signal, the specified The in-vehicle device specified by the occupant of the seat and the operation content thereof are determined. When the determined operation on the determined in-vehicle device is permitted to the occupant of the specified seat, the determined in-vehicle device is determined. A control signal corresponding to the determined operation is output to the device.

According to this configuration, the seat on which the speaker is seated is specified based on the audio signals input from the plurality of microphones, and the content of the audio signal input from the microphone corresponding to the specified seat is determined. The in-vehicle device specified by the occupant of the identified seat and the operation content thereof are determined from the content of the recognized and recognized voice signal, and the determined operation for the identified in-vehicle device is determined by the occupant of the identified seat. And a control signal corresponding to the operation is output to the in-vehicle device. As a result, even when a plurality of occupants are in the vehicle, each occupant can appropriately operate the in-vehicle device by voice.

According to a second aspect of the present invention, there is provided a voice input device for a vehicle for recognizing the content of voice uttered by a vehicle occupant and inputting a control signal to an on-vehicle device. A plurality of microphones arranged corresponding to, a speaker specifying means for specifying a seat where a speaker is seated based on audio signals input from the plurality of microphones, and a plurality of microphones specified by the speaker specifying means. Voice recognition means for recognizing the content of the voice signal input from the microphone corresponding to the seat, wherein the voice recognition means determines the on-board A device and its operation contents are determined, control contents for the determined in-vehicle device are determined according to the specified seat, and a control signal corresponding to the determined control contents is determined. And characterized in that the output.

According to this configuration, the seat on which the speaker is seated is specified based on the audio signals input from the plurality of microphones, and the content of the audio signal input from the microphone corresponding to the specified seat is determined. The in-vehicle device specified by the occupant of the identified seat and the operation content thereof are determined from the content of the recognized and recognized voice signal, and the control content for the in-vehicle device is determined according to the identified seat. , A control signal corresponding to the determined control content is output. As a result, even when a plurality of occupants are in the vehicle, each occupant can appropriately operate the in-vehicle device by voice.

According to a third aspect of the present invention, in the vehicle voice input device according to the first aspect, the voice recognition means includes:
When the specified seat is the driver's seat, execution of the operation of the in-vehicle device related to the driving operation of the vehicle is permitted, and when the specified seat is a seat other than the driver's seat, The operation of the in-vehicle device related to the driving operation is not executable.

According to this configuration, when the specified seat is the driver's seat, execution of the operation of the on-vehicle device related to the driving operation of the vehicle is permitted, and the specified seat is a seat other than the driver's seat. In some cases, the operation of the in-vehicle device related to the driving operation of the vehicle is disabled. As a result, it is possible to prevent the in-vehicle device related to the driving operation from being operated against the driver's intention.

According to a fourth aspect of the present invention, in the vehicle voice input device according to the second aspect, the voice recognition means comprises:
According to the specified seat, a control signal for operating an in-vehicle device corresponding to the seat is output.

According to this configuration, a control signal for operating the vehicle-mounted device corresponding to the specified seat is output according to the specified seat. As a result, more appropriate control can be performed according to the occupant who has given the operation instruction by voice.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of a main part of a vehicle voice input device according to a first embodiment of the present invention. As shown in FIG. 4, this device includes a driver-side microphone 1A and an assistant-side microphone arranged on a shoulder belt portion 10 that restrains the occupant's chest in a driver's seat and a passenger's seat when wearing the seat belt for protecting the occupant. 1B and an audio signal input from the microphones 1A and 1B, a seat in which the speaker is seated is specified, and an audio signal SOU from the microphone corresponding to the specified seat is specified.
T, and an audio signal adjustment unit 2 that outputs speaker information SPK indicating the specified seat. The audio signal adjusting unit 2 recognizes the content of the audio signal SOUT, and sends a control signal CT to the in-vehicle device according to the speaker information SPK.
And a voice recognition unit 3 for outputting L. Examples of in-vehicle devices to be operated by voice include a lamp, a wiper, a mirror, a power window, an air conditioner, a car audio, a navigation device, an auto cruise device, and the like. In the present embodiment, the microphones 1A and 1B are configured so that their characteristics (sensitivity and directivity) are substantially the same.

The voice adjusting unit 2 executes a process as shown in FIG. 2, and outputs an output signal SOUT and speaker information SPK. In step S1, it is determined whether or not a voice signal corresponding to the voice uttered by the driver has been input from the microphone 1A. If there is no voice input from the microphone 1A, the process waits. If there is, the assistant from the microphone 1B. It is determined whether or not a voice signal corresponding to a voice uttered by an occupant sitting in the seat (hereinafter referred to as “assistant”) has been input (step S2). The determination in steps S1 and S2 is performed, for example, as follows.

Input signals S from the microphones 1A and 1B
The levels of IN1 and SIN2 are set to LSIN1 and LSIN2.
Then, the past 30 of the levels LSIN1 and LSIN2
Second moving average values LAVE1 and LAVE2 are calculated, and a threshold TH1 of the input signal SIN1 and a threshold TH2 of the input signal SIN2 are calculated by the following equations.

TH1 = LAVE1 + ΔL1 TH2 = LAVE2 + ΔL2 Here, ΔL1 and ΔL2 are moving average values LAV, respectively.
This is an addition term set to a value of about 5% of E1 and LAVE2.

When LSIN1> TH1 is satisfied, it is determined that there is a driver-side voice input (input from the microphone 1A). When LSIN2> TH2 is satisfied, there is a assistant-side voice input (input from the microphone 1B). Determine. As a result, in a state where only a normally input audio signal, for example, background noise such as road noise is input, it is determined that there is no audio input.

When the answer in step S1 is affirmative (YES) and the answer in step S2 is negative (NO), that is, when only the voice input from the microphone 1A is made, the speaker is the driver (the speaker is in the driver's seat). Seated occupant)
The speaker information SPK is set to "1" (step S4).
At the same time, the audio signal from the microphone 1A is set as the output signal SOUT (step S5), and the output signal SOUT and the speaker information SPK are output to the voice recognition unit 3 to perform step S5.
Return to 1.

On the other hand, if the answers of steps S1 and S2 are both affirmative (YES), that is, if voice input is made from both microphones 1A and 1B, the level LSIN1 of input signal SIN1 from microphone 1A is changed to the input level from microphone 1B. It is determined whether or not the value is greater than a value obtained by adding a predetermined value ΔL3 to the level LSIN2 of the signal SIN2 (step S3). If LSIN1> LSIN2 + ΔL3, the utterer is determined to be a driver, and the process proceeds to the step S
Proceed to 4. When LSIN1 ≦ LSIN2 + ΔL3, it is determined whether or not the level LSIN2 is greater than a value obtained by adding a predetermined value ΔL3 to the level LSIN1 (step S6). And speaker information SP
K is set to "2" (step S7), and an audio signal from the microphone 1B is set as an output signal SOUT (step S8), and the process proceeds to step S9.

If the answers of steps S3 and S6 are both negative (N
O), when the levels LSIN1 and LSIN2 are almost the same, it means that the driver and the assistant have uttered the same volume, or that the voice has been suddenly input from the outside, Since the speaker cannot be specified, the process immediately returns to step S1, and the output signal SOUT and the speaker information are not output.

The predetermined value ΔL3 is set to, for example, a value of about 5% of an average audio signal level at the time of utterance.

FIG. 3 is a flowchart of a process in the voice recognition unit 3. This process is executed when the signal SOUT is input from the voice adjustment unit 2.

First, the contents of the uttered voice are recognized based on the signal SOUT (step S11), and then the in-vehicle device to be operated by the speaker and the operation contents are specified. For example, on-vehicle devices related to driving operations such as lamps, wipers, mirrors, and auto cruise devices, turn on / off the lamps, turn on / off the wipers, and move the fender mirrors (the position between the parking position and the standing position during use). Operation), the operation details such as turning on / off the auto cruise. In addition, turning on / off or controlling the temperature of the air conditioner that is another in-vehicle device, opening and closing the power window,
An instruction to change the display screen of the display unit of the navigation device is specified.

Then, it is determined whether or not the target device and the operation specified in step S12 are permitted by the speaker (step S13). This determination is made by comparing the operation permission table stored in the memory in advance with the detected target device and operation. The operation permission table is
For example, it is set as follows.

1) Operations on on-vehicle devices related to driving operations are permitted only when the speaker is a driver.

2) The navigation device is a target device,
The operation of changing the display screen of the display unit is permitted only when the speaker is an assistant while the vehicle is running. This is because, when the driver can change the display screen during driving, the user is distracted by the screen, and the driver's attention to the front may be reduced.

If the answer to step S13 is negative (NO), the process ends without outputting the control signal CTL for operating the target device.

On the other hand, the answer to step S13 is affirmative (YES).
In the case of, the control content is determined according to the speaker, and a control signal CTL corresponding to the control content is generated. For example, when a navigation device having a main body display unit disposed in the center of an instrument panel or a console as a map display unit and a head-up display projecting on a windshield in front of the driver is to be controlled, a command "map display" is issued. When the driver speaks, a control signal for operating both the main body display unit and the head-up display is generated, and when the same command is issued by the assistant, a control signal for operating only the main body display unit is generated.
Furthermore, opening and closing of the power window is such that when the speaker is a driver, only the driver side window is opened and closed, and when the speaker is an assistant, only the passenger side window is opened and closed, or The details of control such as changing the volume of the car audio only for the signal output from the speaker provided near the speaker's seat are determined, and the corresponding control signal CT is determined.
Generate L. For simple operations such as turning on / off the light and turning on / off the air conditioner, the control signal CTL corresponding to the operation is simply generated.

Next, the control signal CTL is output to the target device, and the process ends.

As described above, in the present embodiment, the seat on which the speaker is seated, that is, whether the speaker is a driver or an assistant is specified based on the audio signals input from the microphones 1A and 1B, and the speaker is identified. Recognizes the content of the identified audio signal, and from the content of the recognized audio signal,
Determines the in-vehicle device specified by the speaker and the contents of the operation, and outputs a corresponding control signal to the in-vehicle device when the determined operation is permitted for the identified in-vehicle device. Therefore, even when the assistant is riding in addition to the driver, each of them can appropriately operate the in-vehicle device by voice.

Further, since the details of the control contents are determined according to the speaker's seat, for example, opening and closing of a window and volume adjustment of a speaker can be performed more appropriately.

Further, since the operation on the on-vehicle device related to the driving operation is permitted only when the driver utters, the in-vehicle device related to the driving operation is inconsistent with the driver's intention by the voice instruction of the assistant. Operation can be prevented.

In the present embodiment, the voice signal adjusting unit 2 corresponds to a speaker specifying unit, and the voice recognizing unit 3 corresponds to a voice recognizing unit.

The present invention is not limited to the embodiment described above, and various modifications are possible. For example, the number of microphones is not limited to two. For example, three or more microphones may be arranged corresponding to a rear seat. The microphone is not limited to be installed at the shoulder belt portion. For example, an arm extending from the headrest near the mouth of the occupant seated on the headrest may be provided and arranged on the arm.

[0036]

As described in detail above, claim 1 or 2
According to the invention described in the above, the seat on which the speaker is seated is specified based on the audio signals input from the plurality of microphones, and the content of the audio signal input from the microphone corresponding to the specified seat is recognized. Then, from the content of the recognized voice signal, the in-vehicle device specified by the occupant of the specified seat and the operation content thereof are determined, and the control signal corresponding to the specified seat, the operation target device, and the operation content is output. Therefore, even when a plurality of occupants are in the vehicle, each occupant can appropriately operate the in-vehicle device by voice.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a vehicle voice input device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a process in an audio signal adjustment unit in FIG. 1;

FIG. 3 is a flowchart illustrating a process in a voice recognition unit in FIG. 1;

FIG. 4 is a diagram for explaining the arrangement of microphones.

[Explanation of symbols]

 1A, 1B microphone 2 Voice signal adjustment unit (speaker identification unit) 3 Voice recognition unit (voice recognition unit)

Claims (4)

[Claims] 1. A vehicular voice input device for recognizing voice content uttered by an occupant of a vehicle and inputting a control signal to an on-vehicle device, wherein the voice input device is arranged corresponding to a plurality of seats of the vehicle. A plurality of microphones, a speaker identification unit that identifies a seat where the speaker is seated based on audio signals input from the plurality of microphones, and a microphone corresponding to the seat identified by the speaker identification unit. Voice recognition means for recognizing the content of the input voice signal, wherein the voice recognition means determines on-board equipment specified by the occupant of the specified seat and its operation content from the content of the recognized voice signal. Then, when the determined operation on the determined in-vehicle device is permitted to the occupant of the specified seat, the determined in-vehicle device is Vehicle voice input device comprising outputting a control signal corresponding to the serial was determined operation. 2. A vehicle voice input device for recognizing the content of voice uttered by an occupant of a vehicle and inputting a control signal to on-vehicle equipment, wherein the voice input device is arranged corresponding to a plurality of seats of the vehicle. A plurality of microphones, a speaker identification unit that identifies a seat where the speaker is seated based on audio signals input from the plurality of microphones, and a microphone corresponding to the seat identified by the speaker identification unit. Voice recognition means for recognizing the content of the input voice signal, wherein the voice recognition means determines on-board equipment specified by the occupant of the specified seat and its operation content from the content of the recognized voice signal. And determining a control content for the determined in-vehicle device according to the specified seat, and outputting a control signal corresponding to the determined control content. Vehicle for the voice input device to be. 3. The voice recognition means, when the specified seat is a driver's seat, permits execution of an operation of an in-vehicle device related to a driving operation of the vehicle, and the specified seat is a driver's seat. 2. The vehicle voice input device according to claim 1, wherein when the seat is a seat other than the vehicle, the operation of the in-vehicle device related to the driving operation of the vehicle is disabled. 3. 4. The vehicle voice input according to claim 2, wherein the voice recognition unit outputs a control signal for operating an on-vehicle device corresponding to the specified seat according to the specified seat. apparatus.