KR100331574B1 - An Automatic Sunroof Controller and A Method Use Of Speech Recognition - Google Patents

Abstract

The present invention relates to a sunroof opening and closing system of a vehicle, and more particularly, by automatically opening and closing the sunroof of a vehicle using voice recognition technology, which may occur due to the neglect of the front gaze by the opening and closing operation of the sunroof while driving. The present invention relates to a vehicle sunroof automatic control apparatus and method using a voice recognition technology to prevent a safety accident in advance.

In the conventional automatic sunroof, since the control switch is mainly located at the front of the sunroof, the operation can sometimes be inconvenient during operation, and with one switch, 'Open, Close, Tilt-on, Tilt-off' , One step Down ... ', so the driver has to check the eye up, there was a dangerous problem that can cause a safety accident.

The present invention to solve the above problems,

A preamplifier for amplifying to improve the signal noise ratio of the minute signal input through the microphone;

A speech recognition processor for controlling the entire system operation according to the speech signal amplified by the preamplifier;

A gain adjuster for adjusting the gain of the preamplifier under the control of the speech recognition processor;

A digital / analog converter configured to convert a digital signal for voice command training into an analog signal under the control of the voice recognition processor;

An amplifier for amplifying the analog signal converted by the digital / analog converter;

A speaker for outputting an announcement sound as an audible sound so that a user can listen to the amplified signal from the amplifier and train a voice command;

A driver for driving the relay on / off under the control of the voice recognition processor;

A sun loop which is automatically controlled according to the driving of the relay; And,

A speaker selection switch for selecting a voice commander for automatically controlling the sunroof by voice,

A first step of determining whether a startup key has been entered after initializing the system;

A second step of waiting for a startup key to be input when the startup key is not input in the first step, and performing a startup key processing if the startup key is input;

Performing a start key process in the second step, and then pressing a switch to generate an operation sound;

A fourth step of determining whether to recognize during operation when the switch operation sound is generated once in the third step;

A fifth step of feeding back to the first step when not in recognition in the fourth step and performing a recognition process when recognizing;

A sixth step of judging whether the switch operation sound is generated twice in the third step;

A seventh step of performing a speaker selection process when the voice trained in the sixth step; And,

In the case where the voice trained in the sixth step is not the sixth step, the training process is performed.

Therefore, according to the present invention, the driver can conveniently control the operation of the sunroof with a voice command without being disturbed to drive, thereby providing the driver with the convenience of operation. By preventing the safety accidents that may occur in advance, there is an effect that can provide safety of operation.

Description

An Automatic Sunroof Controller and A Method Use Of Speech Recognition}

The present invention relates to a sunroof opening and closing system of a vehicle, and more particularly, by automatically opening and closing the sunroof of a vehicle using voice recognition technology, which may occur due to the neglect of the front gaze by the opening and closing operation of the sunroof while driving. The present invention relates to a vehicle sunroof automatic control apparatus and method using a voice recognition technology to prevent a safety accident in advance.

In general, the sun roof (SUN LOOP) is installed in the ceiling portion of the car is widely used in the advanced countries of the country, and is also increasing in our country.

Automatic sunroof can be inconvenient during operation because control switch is mainly located in front of sunroof, and with one switch, 'Open, Close, Tilt-on, Tilt-off, One step' Down ... ', so the driver has to check the eye up, there was a dangerous problem that can cause a safety accident.

Accordingly, the present invention was devised to solve the above-mentioned problems, and by opening and closing the sunroof of the car automatically using voice recognition technology, the neglect of the front gaze by the opening and closing operation of the sunroof while driving the car. It is an object of the present invention to provide an apparatus and method for automatically controlling a sunroof of a vehicle using voice recognition technology, which can prevent accidents that may occur in advance.

The sunroof automatic control device for a vehicle using the voice recognition technology of the present invention for achieving the above object,

A preamplifier amplified to improve the signal noise ratio of the minute signal input through the microphone;

A speech recognition processor for controlling the entire system operation according to the speech signal amplified by the preamplifier;

A gain adjuster for adjusting the gain of the preamplifier under the control of the speech recognition processor;

A digital / analog converter configured to convert a digital signal for voice command training into an analog signal under the control of the voice recognition processor;

An amplifier for amplifying the analog signal converted by the digital / analog converter;

A speaker for outputting an announcement sound as an audible sound so that a user can listen to the amplified signal from the amplifier and train a voice command;

A driver for driving the relay on / off under the control of the voice recognition processor;

A sun loop which is automatically controlled according to the driving of the relay; And,

And a speaker selection switch for selecting a voice commander for automatically controlling the sunroof by voice.

In addition, the sunroof automatic control method of the vehicle using the voice recognition technology of the present invention for achieving the above object,

A first step of determining whether a startup key has been entered after initializing the system;

A second step of waiting for a startup key to be input when the startup key is not input in the first step, and performing startup key processing if the startup key is input;

Performing a start key process in the second step, and then pressing a switch to generate an operation sound;

A fourth step of determining whether to recognize during operation when the switch operation sound is generated once in the third step;

A fifth step of feeding back to the first step when not in recognition in the fourth step and performing a recognition process when recognizing;

A sixth step of judging whether the switch operation sound is generated twice in the third step;

A seventh step of performing a speaker selection process when the voice trained in the sixth step; And,

In the case where the voice trained in the sixth step is not the sixth step, the training process is performed.

1 is a block diagram of a sunroof automatic control apparatus for a vehicle using a voice recognition technology according to the present invention,

2 is an operation flowchart of a sunroof automatic control method of a vehicle using a voice recognition technology according to the present invention.

<Description of the symbols for the main parts of the drawings>

1: microphone 2: preamplifier

3: voice recognition processor 4: gain adjuster

5: digital to analog converter 6: amplifier

7: speaker 8: driver

9: relay 10: sunroof

11: speaker selector switch

Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings.

As shown in FIG. 1, a pre-amplifier 2 for amplifying the signal noise ratio of a small signal input through the microphone 1 is improved, as shown in FIG. 1. Wow ; A voice recognition processor (3) for controlling the entire system operation according to the voice signal amplified from the preamplifier (2); A gain adjuster (4) for adjusting the gain of the preamplifier (2) under the control of the speech recognition processor (3); A digital-to-analog converter 5 for converting a digital signal for voice command training into an analog signal under the control of the voice recognition processor 3; An amplifier (6) for amplifying the analog signal converted by the digital / analog converter (5); A speaker (7) for outputting an announcement sound as an audible sound so that a user can listen to the amplified signal from the amplifier (6) and train a voice command; A driver 8 for driving the relay 9 on / off under the control of the voice recognition processor 3; A sunroof 10 which is automatically controlled according to the driving of the relay 9; And a speaker selection switch 11 for selecting a voice commander for automatically controlling the sunroof 10 by voice.

The sunroof automatic control apparatus for a vehicle using the voice recognition technology of the present invention configured as described above first checks whether a boot-up signal is received when the driver turns on the vehicle, and the boot-up signal is When coming in, press the desired speaker selection switch (11).

Press the speaker selection switch 11 as described above to enter the recognition mode.

Here, Speech Recognition is a technology that makes it easy for a machine or computer to know what a person has said. It is important for communication from human to machine among man-machine interfaces. Means.

When a computer or machine transmits information through a human voice, a machine must be able to speak, and this technique is called speech synthesis.

Speech recognition is classified into Isolated Word, Connected Word, Word Spotting, and Continuous Speech Recognition according to the recognition object, and Speaker Dependent depending on whether or not the speaker is dependent. ), It is classified as Speaker Independent.

Isolated word recognition is to recognize words or phrases that are generated singly, such as 'one', 'two', and 'three' numbers, and linked words can be connected to words spoken singly, such as 'one, two, three'. The word spotting is a technology for recognizing only the corresponding word among consecutive utterances, and is a preliminary technology of continuous sound recognition for recognizing sentences freely spoken without word restrictions.

In the case of continuous speech recognition, it is necessary to know structural aspects such as grammar of language, so that natural language processing becomes an essential element and its implementation is very difficult technically.

The speaker dependent recognition system means that only the utterance of a specific speaker is guaranteed, and that the speaker independent recognition system can recognize all the utterances of any speaker.

Therefore, since the description of the speaker independent recognition system is technically difficult compared to the speaker dependent recognition system, the speaker dependent recognition system is used in the present invention.

As described above, in the case of speaker-dependent recognition system, training is required, and the user should be able to easily train in case of factors such as fluctuation in recognition rate during use.

Also, in certain circumstances (very noisy, snowy or rainy), it may be advantageous to stop the voice recognition function, so that the voice recognition function should be designed to stop.

On the other hand, when training his or her voice, the speaker selection switch 11 is pressed for 2 seconds or longer to enter the voice command training mode.

At this time, if his voice is not recorded, the speaker selection switch 11 is lightly pressed to enter the training mode even when the speaker is selected.

When entering the training mode, recording six commands using the microphone 1 in accordance with the announcement broadcast through the speaker (7), and automatically enters the recognition mode.

On the other hand, it is possible to press the two of the speaker selection switch 11 at the same time for more than 2 seconds to end the recognition.

The recognition mode is a state in which a user's voice command is waiting for continuous listening.

Here, in the sunroof control by the switch operation, since the two switches are in one socket, various operations can be performed with these two switches. Let's call these switches 1 and 2.

First, the sunroof opening is opened by pressing the switch 1 in a state in which the sunroof 10 is closed. When opened, pressing any switch of Switch 1 or Switch 2 opens and stops the sunroof 10, and opens to the end if there is no switch operation in the middle.

The sunroof closing is closed by pressing switch 2 while the sunroof 10 is open. When the door is closed, the sunroof 10 stops when one of the switches 1 and 2 is pressed in the same way as when the door is opened.

On the other hand, there is also a function that causes the sunroof 10 to remember the open position so that the sunroof 10 is opened only to the stored position the next time the sunroof 10 is opened.

This is because when the sunroof 10 is closed, the switch 1 is opened, the sunroof 10 is stopped at the desired position, and the switch 2 is opened after about 2 seconds even after the sunroof 10 is closed. Hold down.

Then, when the sunroof 10 memorizes the position and opens the sunroof 10 again, only the stored position is opened.

The release method may be a position memory in a state where the sunroof 10 is closed or completely open.

In the tilt mode, the opening and closing function opens only the back of the sunroof 10 by pressing the switch 2 while the sunroof 10 is closed.

If you press the switch 1 in the open state, it is closed again. If you press and hold the switch 1 in the open state, it is divided into 5 stages. When you release the switch 1 from the desired position, it stops at that position.

As described above, six voice commands are required to control the sunroof control function that can be controlled by a switch.

The six voice commands are sunroof open, sunroof close, sunroof stop, sunroof up, sunroof close, tilt off, and tilt. In the open mode, it can be classified into 5 steps (Step Down).

In the method of controlling the sun loop using the voice command as described above, first, the voice command input through the microphone 1 receives an electrical signal to the driver 8 when the voice command is recognized by the voice recognition processor 3. The relay 9 is operated to control the sunroof 10.

However, in this case, the voice command input through the microphone 1 is wrong, but since the electrical signal output from the voice recognition processor 3 to the driver 8 after recognition is the same, the same operation occurs even if other commands are spoken.

That is, a sunroof open command is issued, and the sunroof close or sunroof stop is executed during the sunroof 10 opening. Since the relay 9 is controlled by a stop or the same electrical signal, the sunroof 10 is stopped.

In order to solve the problem of speech mismatch, it is necessary to limit the operation in software.

It is to distinguish between commands that can follow each voice command and commands that cannot follow.

In the above case, for example, after the sunroof open, the voice recognition processor 3 is restricted from operating other than the sunroof stop.

The voice recognition processor 3 issues a voice command, but there must be a command allowed according to the state of the sunroof 10 and a command prohibited.

For example, when the sunroof 10 is opened with sunroof open, the sunroof open command or the step down command is not allowed when the sunroof is open in tilt mode. Should not.

Thus, logic is required to distinguish between commands that can follow each voice command and commands that cannot follow.

2 illustrates a first step (S1) of determining whether a startup key is input after initializing the system as shown in FIG. 2; A second step (S2) of waiting for a startup key to be input when the startup key is not input in the first step (S1) and performing a startup key processing if the startup key is input; A third step (S3) for generating an operation sound by pressing a switch after performing the start key processing in the second step (S2); A fourth step (S4) of determining whether the recognition operation is occurring when the switch operation sound is generated once in the third step (S3); A fifth step (S5) of feeding back to the first stage (S1) system if not in recognition in the fourth step (S4), and performing recognition processing if it is in recognition; A sixth step S6 of determining whether a trained voice is generated when the switch operation sound is generated twice in the third step S3; A seventh step (S7) of performing a speaker selection process when the voice trained in the sixth step (S6); And an eighth step S8 of performing a training process when the voice trained in the sixth step S6 is not performed.

As described in detail above, the apparatus and method for automatically controlling a sunroof of a vehicle using a voice recognition technology according to the present invention provide convenience to the driver by controlling the operation of the sunroof simply by a voice command without being disturbed by the driver. It can be provided, thereby preventing the safety accidents that may occur due to the neglect of the front gaze due to the opening and closing operation of the sunroof during operation has the effect of providing the safety of the operation.

Although the invention has been described in detail only with respect to the specific examples described, it will be apparent to those skilled in the art that various modifications and variations are possible within the spirit of the invention, and such modifications and variations belong to the appended claims.

Claims (2)

A preamplifier (2) for amplifying to improve the signal noise ratio of the minute signal input through the microphone (1); A voice recognition processor (3) for controlling the entire system operation according to the voice signal amplified from the preamplifier (2); A gain adjuster (4) for adjusting the gain of the preamplifier (2) under the control of the speech recognition processor (3); A digital-to-analog converter 5 for converting a digital signal for voice command training into an analog signal under the control of the voice recognition processor 3; An amplifier (6) for amplifying the analog signal converted by the digital / analog converter (5); A speaker (7) for outputting an announcement sound as an audible sound so that a user can hear a signal amplified by the amplifier (6) and train a voice command; A driver 8 for driving the relay 9 on / off under the control of the voice recognition processor 3; A sunroof 10 which is automatically controlled according to the driving of the relay 9; And, A sunroof automatic control device for a vehicle using a voice recognition technology, characterized in that the speaker selection switch for selecting a voice commander for the automatic control of the sunroof (10) by voice. A first step S1 of determining whether a ignition key has been input after initializing the system; A second step (S2) of waiting for a startup key to be input when the startup key is not input in the first step (S1) and performing a startup key processing if the startup key is input; A third step (S3) for generating an operation sound by pressing a switch after performing the start key processing in the second step (S2); A fourth step (S4) of determining whether the recognition operation is occurring when the switch operation sound is generated once in the third step (S3); A fifth step (S5) of feeding back to the first stage (S1) system if not in recognition in the fourth step (S4), and performing recognition processing if recognition is in progress; A sixth step S6 of determining whether a trained voice is generated when the switch operation sound is generated twice in the third step S3; A seventh step (S7) of performing a speaker selection process when the voice trained in the sixth step (S6); And, If it is not the voice trained in the sixth step (S6), the sunroof automatic control method of the vehicle using the voice recognition technology, characterized in that it comprises an eighth step (S8) of performing a training process.