TWI749797B - Sound source capturing device of driving alert and method thereof - Google Patents

Abstract

The present invention relates a sound source capturing device of driving alert which utilizes a radio module receives a long-distance sound signal in a long-distance range from the outside of vehicle, and a processor generates a warning sound source through amplifying a sound feature point audio in a sound source accords to warning voiceprint information and generates a plurality of adjusted sound sources through reducing or shielding the sound feature point audio in the other sound sources accords to non-warning voiceprint information. Then the processor combines the warning sound source and the plurality of adjusted sound sources to generate an output sound signal, allowing an output module to output the output sound signal. Accordingly, the capturing device of the present invention provides instantly warning sounds receiving (e.g. sound of car engine) from a long-distance range and outputs to allow the driver to be early alert state, reducing the probability of incident occurrence thereby.

Description

Device and method for detecting sound source of driving warning

The present invention relates to the technical field of sound source detection, in particular to a driving warning sound source detection device and method for detecting related warning sound sources in a remote range outside the vehicle body.

According to this, in the environment of a driving vehicle, because the surrounding environmental sound or other sounds easily overwhelm the sound of a specific object (for example, other vehicles) from the remote end, it is difficult for driving in this environment to hear the sound of the specific object, so, When the driving route and the driving route of a specific object may overlap or be too close, accidents may occur because either party is not paying attention or the driving/specific object is driving too fast, such as a car (specific Subject) When rushing out of the alley, a driver driving on a straight road could not immediately notice a car coming out of the alley.

In this case, even if a radio cover provided by the prior art is installed on a radio device, the radio device has to focus on receiving the sound emitted in one direction through physical constriction.

However, the warning sound source (such as car engine sound, locomotive engine sound, ambulance, police car and fire engine horn sound, etc.) emitted at the remote end is usually transmitted from all directions instead of being uploaded in a specific direction. The prior art cannot effectively provide driving warnings when it only receives sound in a specific direction The role. Furthermore, the radio device cannot automatically identify the warning sound source, and can only provide the driver to select a specific object when the radio device receives a plurality of sound sources, and then amplify the sound source of the specific object according to related processing methods.

Since the speed of driving reaction is extremely important before the danger occurs, if the previous technology is used to deal with it, it may not have time to provide driving-related warning sound sources, and related accidents have already occurred.

Therefore, there is an urgent need for a technology that can automatically and effectively identify warning sound sources emitted from a remote range from all directions, so as to improve the problems of the prior art.

The object of the present invention is to provide a sound source detection device for driving warning, which mainly uses the remote sound signal sensed from a remote range outside the car body, and when at least one sound source separated from the remote sound signal conforms to When warning voiceprint data, the processing module amplifies the sound feature point audio in the sound source, and then generates a warning sound source to output to the driver to listen, so as to provide real-time output of the warning sound from the remote end, which facilitates early warning for driving and reduces accidents The probability of occurrence, thereby improving the problems of the previous technology.

For the purpose of the above disclosure, the present invention provides a sound source detection device for driving warning, which includes: a body, which is arranged on a vehicle body; a radio module, which is arranged in the body and from the vehicle Receive a remote sound signal in a remote range outside the body; a processing module is set in the body and connected with the radio module to receive the remote sound signal, and separate the remote sound signal A plurality of sound sources, the processing module compares a sound feature point in the sound sources based on a plurality of warning voiceprint data and a plurality of non-alert voiceprint data, and the processing module uses a deep learning algorithm to extract Among the sound feature points, a sound feature point audio of the sound source that matches at least one of the warning voiceprint data is amplified, and the sound feature point audio in the sound source that matches the warning voiceprint data is amplified to generate a warning sound source, and Reducing or shielding the sound feature point audio in the sound sources that meet at least one of the non-alert voiceprint data to generate a plurality of adjusted sound sources, wherein the processing module executes a synthesis process for the warning sound source and the adjusted sound sources , So that the warning sound source and the adjusted sound sources are combined to generate an output sound signal; and an output device, which is arranged in the body and connected with the processing module, and the output device receives and outputs the output sound signal.

Preferably, the processing module amplifies the sound feature point audio in the sound source that matches the warning voiceprint data according to an amplification threshold.

Preferably, the sound source detection device for driving warning further includes: a detection module connected to the radio module, the detection module provides a sound wave in the remote range, and senses the sound wave according to the sound wave The vibration waveform generates the remote sound signal, and the detection module sends the remote sound signal to the radio module.

Preferably, the distal range includes 1 meter to 10 meters.

Preferably, the processing module transmits the output sound signal to the output device that matches the direction in the vehicle body according to a direction in which the radio module receives the remote sound signal, and the The output device outputs the output sound signal.

Preferably, the sound source detection device for driving warning further includes: a register, which is connected with the processing module to store the warning voiceprint data, the non-warning voiceprint data, the remote sound signal, the The sound feature points, the sound feature point audio, the output sound signal, or a combination of two or more thereof.

Preferably, the sound source detection device for driving warning further includes: an activation module connected to the processing module and the radio module, the activation module receives the remote sound signal from the radio module, and according to An activation threshold is compared with a sound source data of the remote sound signal to determine whether to enable The processing module is activated. When the sound source data meets the activation threshold, the activation module activates the processing module to make the processing module separate the sound sources in the remote sound signal. When the sound source data is not When the activation threshold is met, the activation module continues to receive another remote audio signal.

Preferably, the sound source data includes a decibel value, a frequency value or a combination of the two.

Preferably, the processing module is connected to a vehicle control device to receive a control signal from the vehicle control device, and the processing module controls the radio module or the output device according to at least one command in the control signal.

Another object of the present invention is to provide a method for detecting a remote warning sound source, which is mainly used to sense the remote sound signal in a remote range outside the car body, and when the remote sound signal is separated at least When a sound source matches the warning voiceprint data, the processing module amplifies the sound feature point audio in the sound source, and then generates a warning sound source for output to be heard by the driver, so as to provide real-time output of the warning sound from the remote end, which is beneficial for early warning of the driver. In order to reduce the probability of accidents, and then improve the problems existing in the prior art.

In order to achieve another objective of the above disclosure, the present invention provides a method for detecting the sound source of driving warning applied to the sound source detecting device of driving warning as described above.

In order to make the above-mentioned objectives, features and advantages of the present invention more obvious and understandable, detailed descriptions are given below in conjunction with the specific embodiments listed in the drawings.

100: body

200: Vehicle control device

201: Control signal

10: Radio module

11: Remote audio signal

111: Sound source data

20: Processing module

21: Sound source

211: Sound Feature Points

212: Sound feature point audio

22: Warning voiceprint data

23: Non-alert voiceprint data

24: Warning sound source

25: Adjust the sound source

26: Enlarge the threshold

27: Output audio signal

30: output device

40: Exploring Module

41: Sonic

50: register

60: Activate the module

61: Activation threshold

S01-S08: Step process

Figure 1 is a schematic diagram of the component arrangement relationship of the present invention; Figure 2 is a schematic diagram of the component arrangement relationship of the vehicle control device of the present invention; Figure 3 is a schematic diagram of the activation module configuration of the present invention; Figure 4 is a flow chart of the steps of the present invention.

The advantages, features, and technical methods of the present invention will be described in more detail with reference to exemplary embodiments and the accompanying drawings to make it easier to understand, and the present invention can be implemented in different forms, so it should not be understood as the present invention It is limited to the embodiments set forth herein. On the contrary, for those with ordinary knowledge in the technical field, the provided embodiments will make this disclosure more thorough, comprehensive and complete to convey the scope of the present invention, and the present invention will only It is defined by the scope of the attached patent application.

In addition, the terms "including" and/or "including" refer to the existence of the features, regions, wholes, steps, operations, elements, and/or components, but do not exclude one or more other features, regions, wholes, steps, operations , The presence or addition of elements, components, and/or combinations thereof.

In order to make it easier for your reviewer to understand the content of the present invention and the effects that can be achieved, the specific embodiments listed in the drawings are described in detail as follows: Please refer to Figures 1 and 2, which are elements of the present invention. The schematic diagram of the configuration relationship and the schematic diagram of the component configuration relationship of the vehicle control device. As shown in the figure, the present invention is to install a main body 100 on a vehicle body, and it is mainly composed of a radio module 10, a processing module 20 and an output device 30 in the main body 100. The main body 100 can be a plastic shell or other container that can accommodate related modules in the main body 100 to provide protection. The main body 100 can be installed on the vehicle body. More specifically, the main body 100 The 100 series can be installed on the inside or outside of the car body.

The radio module 10 may specifically be a microphone or other related devices that can receive external sound. In order to enable the present invention to effectively receive the sound signal to be analyzed and processed in a remote range, it is disclosed here The radio module 10 is limited to the remote range (for example, between 1 meter and 10 meters) to sense a remote sound signal 11 (for example, the sound frequency is between 0HZ and 20000HZ) The remote sound signal 11), more specifically, the sound source detection device for driving warning of the present invention further includes a detection module 40, which is mainly configured to provide a sound source in the remote range The sound wave 41, and when the sound wave 41 is vibrated by an external sound wave, the detection module 40 can determine that the external sound wave has been sensed, and generate the remote sound signal 11 according to the sensed vibration wave pattern. In this way, When the detection module 40 is connected to the radio module 10, the radio module 10 can receive the remote sound signal 11 from the detection module 40. Wherein, the detection module 40 may be a module provided in the main body 100, or the detection module 40 provided on the radio module 10, where the detection module 40 is only The need to be able to provide sensing of the sound wave 41 in the remote range should all belong to the scope of the patent to be protected by the present invention.

The processing module 20 may specifically be a central processing unit or other data processing device. When the processing module 20 is connected to the radio module 10, it can receive the data received by the radio module 10 The remote audio signal 11, at this time, the processing module 20 will first separate the multiple audio sources 21 (such as the sound of a specific object and other environmental sounds) in the remote audio signal 11. When each of the audio sources 21 is separated After it comes out, in order to effectively distinguish the differences between the sound sources 21, the processing module 20 compares a plurality of warning voiceprint data 22 and a plurality of non-alert voiceprint data 23 to a sound feature point in the sound sources 21 211, and the warning voiceprint data 22 and the non-alert voiceprint data 23 can be obtained through a period of sound learning or data stored in an original register 50, wherein the warning voiceprint data 22 Mainly refers to the sound of related objects that may cause accidents or harm to driving, such as car engine sound, locomotive engine sound, ambulance, police car and fire engine horn sound, etc.; on the contrary, these non-warning voiceprint data It refers to the sound of related objects that will not cause accidents or harm to the human body, such as the sound of wind, the sound of doors, or other environmental sounds.

After the sound feature points 211 of the sound sources 21 are compared, the processing module 20 can use a deep learning algorithm to extract the sound feature points 211 to distinguish them as matching at least one of the alarms. Shows the sound source of the voiceprint data 22 and the sound sources 21 that match at least one of the non-alert voiceprint data 23, and after extracting a sound feature point audio 212 in the sound sources 21, the processing module 20 further amplifies and matches The sound feature point audio 212 in the sound source 21 of the warning voiceprint data 22 is used to generate a warning sound source 24 and reduce or shield other sound features in the sound sources 21 that meet at least one of the non-warning voiceprint data 23 Click the audio signal 212 to generate a plurality of adjustment audio sources 25, and then highlight the sound of the warning audio source 24.

Wherein, when the processing module 20 is performing the amplification process, if the sound feature point audio 212 of the warning sound source 24 is over-amplified, it may cause the driving to be frightened and the occurrence of related accidents. In this way, in order to avoid the above In this case, when the processing module 20 executes the amplification procedure, it amplifies the sound feature point audio 212 in the sound source 21 that meets the warning voiceprint data 22 according to an amplification threshold 26, and the amplification threshold 26 is based on a general human body It is set in decibels that will not feel uncomfortable (for example, the voice intensity of ordinary people's speech (50 decibels)).

Thereafter, the processing module 20 executes a synthesis procedure for the warning sound source 24 and the adjustment sound sources 25, so that the warning sound source 24 and the adjustment sound sources 25 are combined to generate an output sound signal 27. In this way, the output sound The signal 27 includes the warning sound source 24 that is highlighted and the adjusted sound sources 25 that are attenuated or muted.

The output device 30 can specifically be a speaker or any related device that can output the output sound signal. After the output sound signal 27 is generated, the output device 30 can receive the output signal by connecting with the processing module 20. The audio signal 27 is output, and the output audio signal 27 is output to the driver to listen. In addition, the processing module 20 can further determine the direction in accordance with the direction in which the radio module 10 receives the remote audio signal 11 The output device 30 receives the output sound signal 27. For example, when the remote sound signal 11 is received from the front left position outside the vehicle body, when the processing After the module 20 completes the action of generating the output sound signal 27, the processing module 20 can separately transmit the output sound signal 27 to the output device 30 installed in the front left position of the vehicle body to receive the remote sound signal 11 , When the driver hears the output sound signal 27, he can understand the direction of the output sound signal 27 at the same time. In turn, it is effective to receive the warning sound from the remote end immediately and output it to the driver to know it, so that the driver can be alerted early to reduce the probability of accidents.

Wherein, the register 50 is connected with the processing module 20 to store the warning voiceprint data 22, the non-alert voiceprint data 23, the remote sound signal 11, the sound feature points 211, the The sound feature point audio 212, the output audio signal 27, or a combination of two or more thereof.

As shown in FIG. 2, the processing module 20 can be further connected to a vehicle control device 200 in the vehicle body to receive a control signal 201 from the vehicle control device 200, and according to at least one command in the control signal 201 Control the radio module 10 or the output device 30. For example, when the driver wants to set (for example, the receiving range) the radio module 10, the vehicle control device 200 can be used to generate the relevant control signal 201 and then the processing module 20 can demodulate the control signal 201 before execution. For related actions, or when the driver wants to adjust the output volume of the output device 30, the control signal 201 generated by the vehicle control device 200 can also be used to perform related operations through the processing module 20.

In addition, since the conversion or processing between digital signals or analog signals is a conventionally known technology, in the above-mentioned signal receiving or outputting actions, the actions known in the prior art will not be repeated.

Please refer to FIG. 3 again, which is a schematic diagram of the activation module configuration of the present invention. As shown in the figure, in order to effectively save power, the sound source detection device for driving warning of the present invention may further include an activation module 60, which is arranged in the main body 100 and interacts with the processing module 20 and the processing module 20 and the The radio module 10 is connected, the activation module 60 receives the remote sound signal 11 from the radio module 10, and according to the activation door The threshold 61 is compared with a sound source data 111 (such as a decibel value, a frequency value or a combination of the two) of the remote sound signal 11 to determine whether to activate the processing module 20, for example, because Both car engine sound and locomotive engine sound have considerable sound intensity and a certain frequency range, so when the sound source data 111 meets the activation threshold 61 (for example, a sound signal with a similar frequency or a sound signal with a decibel intensity is received) When the activation module 60 activates the processing module 20, the processing module 20 separates the audio sources 21 in the remote audio signal 11. When the audio source data 111 does not meet the activation threshold 61, the activation The module 60 continues to receive another remote audio signal 11 to determine the relevant activation procedure. This effectively provides the power-saving function of the sound source detection device for driving warning of the present invention.

Please refer to FIG. 4 again, which is a flowchart of the steps of the present invention. As shown in the figure, the present invention can mainly follow the following steps to achieve the function of detecting the remote warning sound source as described above, which includes: S01: The radio module receives the remote sound at the remote range outside the car body Signal; S02: the processing module separates multiple audio sources in the remote audio signal; S03: the processing module compares the sound feature points in these audio sources based on multiple warning voiceprint data and multiple non-alert voiceprint data; S04: The processing module uses a deep learning algorithm to extract the sound characteristic point audio of the sound source that matches at least one of the warning voiceprint data among the sound characteristic points; S05: The processing module amplifies the sound source that matches the warning voiceprint data S06: The processing module reduces or shields other sound feature point audio in the sound sources that meet at least one of the non-alert voiceprint data to generate a plurality of adjusted sound sources; S07: The processing module executes a synthesis program for the warning sound source and the adjusted sound sources, so that the warning sound source and the adjusted sound sources are combined to generate an output sound signal; S08: the output device receives and outputs the output sound signal.

As a result, the sound source detection device for driving warning of the present invention can immediately receive the warning sound (such as the sound of a car engine) from the remote end, and output it to the driver to learn that it is useful for driving early warning, so as to reduce the probability of accidents. .

The disclosure in this case is a preferred embodiment, and any partial changes or modifications that are derived from the technical ideas of the case and can be easily inferred by those who are familiar with the art will not deviate from the scope of the patent right of this case.

In summary, regardless of the purpose, means, and effects of this case, it is showing its technical characteristics that are very different from conventional knowledge, and its first invention is suitable for practical use, and it is also in compliance with the patent requirements of the invention. Granting patents as soon as possible will benefit the society and feel the virtues.

100: body

10: Radio module

11: Remote audio signal

111: Sound source data

20: Processing module

21: Sound source

211: Sound Feature Points

212: Sound feature point audio

22: Warning voiceprint data

23: Non-alert voiceprint data

24: Warning sound source

25: Adjust the sound source

26: Enlarge the threshold

27: Output audio signal

30: output device

40: Exploring Module

41: Sonic

50: register

Claims (8)

A sound source detection device for driving warning, which includes: a body, which is arranged on a vehicle body; a radio module, which is arranged in the body and receives a remote from a remote range outside the vehicle body Sound signal; a processing module, which is set in the body and connected with the radio module to receive the remote sound signal, and separate the plural sound sources in the remote sound signal, the processing module is based on the plural The warning voiceprint data and a plurality of non-warning voiceprint data are compared to a voice feature point in the sound sources, and the processing module uses a deep learning algorithm to extract at least one of the voice feature points that matches the warning sound Pattern data of a sound feature point audio of the sound source, and amplify the sound feature point audio in the sound source that meets the warning voiceprint data to generate a warning sound source, and reduce or shield other sounds that meet at least one of the non-alert voiceprint data The sound feature point audio in the sound sources is used to generate a plurality of adjusted sound sources, wherein the processing module executes a synthesis process for the warning sound source and the adjusted sound sources, so that the warning sound source and the adjusted sound sources are combined to produce a Output audio signals; an output device, which is provided in the body and is connected to the processing module, the output device receives and outputs the output audio signal; and a register, which is connected to the processing module to store the Warning voiceprint data, the non-alert voiceprint data, the remote voice signal, the voice feature points, the voice feature point audio, the output voice signal, or a combination of two or more thereof; The processing module transmits the output sound signal to the output device that matches the direction in the vehicle body according to a direction in which the radio module receives the remote sound signal, and is output by the output device This output audio signal. The sound source detection device for driving warning according to claim 1, wherein the processing module amplifies the sound feature point audio in the sound source that matches the warning voiceprint data according to an amplification threshold. The sound source detection device for driving warning according to claim 1, further comprising: a detection module connected to the radio module, the detection module provides a sound wave in the remote range, and according to The vibration waveform sensed by the sound wave generates the remote sound signal, and the detection module sends the remote sound signal to the radio module. The sound source detection device for driving warning according to claim 1, wherein the remote range includes 1 meter to 10 meters. The sound source detection device for driving warning according to claim 1, further comprising: an activation module connected to the processing module and the radio module, and the activation module receives the remote from the radio module Acoustic signal, and compare an activation threshold with a sound source data of the remote audio signal to determine whether to activate the processing module. When the sound source data meets the activation threshold, the activation module starts the processing The module enables the processing module to separate the audio sources in the remote audio signal. When the audio source data does not meet the activation threshold, the activation module continues to receive another remote audio signal. The sound source detection device for driving warning according to claim 5, wherein the sound source data includes a decibel value, a frequency value or a combination of the two. The sound source detection device for driving warning according to claim 1, wherein the processing module is connected with a vehicle control device to receive a control signal from the vehicle control device, and the processing module is based on at least one of the control signals Commands control the radio module or the output device. A sound source detection method for driving warning applied to the sound source detection device for driving warning as described in any one of claims 1 to 7, comprising: a radio module in a remote range outside a vehicle body Receive a remote audio signal; a processing module separates the multiple audio sources in the remote audio signal; the processing module compares the multiple audio sources based on the multiple warning voiceprint data and the multiple non-alert voiceprint data A voice feature point; the processing module uses a deep learning algorithm to extract a voice feature point audio of the sound source that matches at least one of the warning voiceprint data among the voice feature points; the processing module amplifies the warning sound The sound characteristic point audio in the sound source of the pattern data is used to generate a warning sound source; the processing module reduces or shields the sound characteristic point audio in the sound sources that meet at least one of the non-alert voiceprint data to generate a plurality of adjustments Audio source; the processing module executes a synthesis procedure for the warning sound source and the adjusted sound sources, so that the warning sound source and the adjusted sound sources are combined to generate an output sound signal; and an output device receives and outputs the output sound signal; One of the registers is connected with the processing module to store the warning voiceprint data, the non-alert voiceprint data, the remote sound signal, the sound feature points, the sound feature point audio, and the output sound Signal or a combination of two or more; wherein the processing module is based on a direction in which the radio module receives the remote sound signal to transmit the output sound signal to the vehicle body that matches the direction An output device, and the output sound signal is output by the output device.

Images