CN115190212A - Call noise reduction method, device, headset device and medium based on headset device - Google Patents

Abstract

The invention discloses a method and device for noise reduction of calls based on earphone equipment, earphone equipment and a computer-readable storage medium. The earphone equipment equipped with an analog microphone and a bone conduction sensor is used to detect digital sound signals according to the bone conduction sensor. Determine whether the wearer of the headset device is outputting voice information; if so, determine whether the wearer is wearing a mask correctly according to the analog signal obtained by collecting the voice information by the analog microphone; after determining that the wearer is wearing the mask correctly , and it is determined that when the wearer makes a voice call based on the headset device, noise reduction is performed according to the first voice signal collected by the bone conduction sensor and the second voice signal collected by the analog microphone. The technical solution of the present invention can accurately detect whether the user wears the mask correctly, and when the user wears the mask, noise reduction can be performed in the process of the user using the earphone device to make a voice call.

Description

Call noise reduction method, device, headset device and medium based on headset device

technical field

The present invention belongs to the technical field of earphone equipment, and in particular, relates to a method and device for noise reduction of calls based on earphone equipment, earphone equipment and a computer-readable storage medium.

Background technique

Nowadays, wearing a mask is an effective way to prevent the rapid spread of the virus on a daily basis. However, on the one hand, there are still many social residents in public places who forget to wear masks and wear masks irregularly, which will undoubtedly increase the risk of virus infection. On the other hand, when people wear masks, due to the shielding effect of the masks on the normal breathing airflow of people, the voices made by people will be dull, and due to the friction between the masks and the skin and/or beards of people's faces Extra noise will be generated, which is more obvious when people wear masks for voice calls, especially when users wear TWS (True Wireless Stereo, true wireless stereo) headset devices to make calls, the sound caused by wearing masks becomes Dull and occasional interfering noise can seriously degrade overall call quality.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to provide a method, device, headset device and computer-readable storage medium for noise reduction of calls based on headset device. The purpose is to accurately detect whether the user is wearing a mask correctly based on the headset device, so as to reduce the noise during the user's voice call using the headset device when it is detected that the user wears a mask to improve the quality of the call.

In order to achieve the above object, the present invention provides a method for noise reduction of calls based on a headset device. The method for noise reduction of calls based on a headset device is applied to a headset device configured with an analog microphone and a bone conduction sensor. Noise reduction methods include:

Determine whether the wearer of the earphone device is outputting voice information according to the digital sound signal detected by the bone conduction sensor;

If so, then determine whether the wearer is wearing the mask correctly according to the analog signal obtained by the analog microphone collecting the voice information;

When it is determined that the wearer is wearing the mask correctly, and it is determined that the wearer makes a voice call based on the headset device, the first voice signal collected by the bone conduction sensor and the second voice collected by the analog microphone are used. Signal for call noise reduction.

Optionally, the step of performing noise reduction for calls according to the first voice signal collected by the bone conduction sensor and the second voice signal collected by the analog microphone includes:

extracting a low-frequency voice signal from the first voice signal collected by the bone conduction sensor when the wearer makes a voice call;

extracting a high-frequency voice signal from the second voice signal collected by the analog microphone when the wearer makes a voice call;

The low-frequency voice signal and the high-frequency voice signal are mixed into one channel of voice data output for noise reduction of calls.

Optionally, the step of determining whether the wearer of the earphone device is outputting voice information according to the digital sound signal detected by the bone conduction sensor includes:

acquiring the digital sound signal detected by the bone conduction sensor, and performing speech recognition on the digital sound signal;

When voice recognition is performed on the digital sound signal to detect voice information from the digital sound signal, it is determined that the wearer of the earphone device is outputting voice information.

Optionally, the step of determining whether the wearer is wearing a mask correctly according to an analog signal obtained by collecting the voice information with the analog microphone, includes:

acquiring an analog signal obtained by collecting the voice information continuously output by the wearer by the analog microphone;

Whether the wearer is wearing the mask correctly is determined according to the proportional relationship between the high-frequency energy and the low-frequency energy in the analog signal.

Optionally, the step of determining whether the wearer is wearing the mask correctly according to the ratio between the high-frequency energy and the low-frequency energy in the analog signal includes:

After performing fast Fourier transform on the analog signal, calculate the high-frequency energy and the low-frequency energy in the analog signal respectively, and determine the ratio relationship between the high-frequency energy and the low-frequency energy;

If the proportion relationship is that the proportion of the low-frequency energy exceeds a preset threshold, it is determined that the wearer is wearing a mask correctly;

If the proportion relationship is that the proportion of the low-frequency energy exceeds a preset threshold, it is determined that the wearer does not wear a mask or does not wear a mask correctly.

Optionally, after the step of determining whether the wearer is wearing the mask correctly according to the analog signal obtained by collecting the voice information with the analog microphone, the method further includes:

When it is determined that the wearer does not wear a mask or does not wear a mask correctly, a prompt voice for reminding to wear a mask properly is output through the speaker of the earphone device.

Optionally, the headset device establishes a Bluetooth connection with the mobile terminal in advance, and the method further includes:

When receiving the voice Bluetooth signal transmitted by the mobile terminal through the Bluetooth connection, it is determined that the wearer is currently making a voice call based on the headset device.

In addition, in order to achieve the above object, the present invention also provides a call noise reduction device based on a headset device, which is applied to a headset device configured with an analog microphone and a bone conduction sensor. The device's call noise reduction features include:

The mask detection module is used to determine whether the wearer of the headset device is outputting voice information according to the digital sound signal detected by the bone conduction sensor; a signal to determine whether the wearer is wearing the mask correctly;

A call noise reduction module, used for determining that the wearer is wearing a mask correctly, and determining that the wearer makes a voice call based on the headset device, according to the first voice signal collected by the bone conduction sensor and the simulation The second voice signal collected by the microphone is used for noise reduction of the call.

Each functional module of the headset device-based call noise reduction device of the present invention implements the steps of the above-mentioned control method for the motion monitoring motion of the wireless headset during operation.

In addition, in order to achieve the above object, the present invention also provides a headset device, the headset device includes: a memory, a processor, and a headset device-based call noise reduction device that is stored in the memory and can run on the processor. The program, when the control program for the motion monitoring of the wireless headset is executed by the processor, implements the steps of the above-mentioned control method for the motion monitoring of the wireless headset.

In addition, in order to achieve the above object, the present invention also provides a computer-readable storage medium on which a call noise reduction program based on a headset device is stored, and the call noise reduction program based on the headset device is processed When the device is executed, the steps of implementing the above-mentioned method for noise reduction of calls based on the headset device.

An earphone device-based call noise reduction method, device, earphone device, and computer-readable storage medium proposed in the embodiments of the present invention, through the earphone device configured with an analog microphone and a bone conduction sensor, according to the digital signal detected by the bone conduction sensor. The sound signal determines whether the wearer of the earphone device is outputting voice information; if so, determine whether the wearer is wearing the mask correctly according to the analog signal obtained by collecting the voice information by the analog microphone; after determining that the wearer is correct When wearing a mask and it is determined that the wearer conducts a voice call based on the headset device, noise reduction is performed according to the first voice signal collected by the bone conduction sensor and the second voice signal collected by the analog microphone.

In this way, the embodiment of the present invention realizes that the bone conduction sensor and the simulated microphone can be used to accurately detect whether the user is wearing the mask correctly, and, in the case of detecting that the user is wearing the mask, the bone conduction sensor and the simulated microphone are further used to detect whether the user is wearing a mask correctly. Noise reduction is performed on the voice signal during the voice call using the headset device, thereby improving the voice quality of the user's call based on the headset device.

Description of drawings

1 is a schematic diagram of a device structure of a hardware operating environment of a headset device involved in an embodiment of the present invention;

FIG. 2 is a schematic flowchart of the first embodiment of the method for noise reduction of calls based on earphone equipment according to the present invention;

3 is a schematic flowchart of mask wearing detection involved in an embodiment of the headset device-based call noise reduction method of the present invention;

4 is a schematic flowchart of a call noise reduction involved in an embodiment of a call noise reduction method based on a headset device of the present invention;

FIG. 5 is a schematic diagram of functional modules of an embodiment of a call noise reduction device based on a headset device of the present invention.

The realization, functional characteristics and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

Referring to FIG. 1 , FIG. 1 is a schematic diagram of a device structure of a hardware operating environment of an earphone device according to an embodiment of the present invention.

The headset device in the embodiment of the present invention may specifically be a headset device such as a Bluetooth headset, a TWS (True Wireless Stereo, true wireless stereo) headset, etc. The headset device is internally equipped with an analog MIC—an analog microphone and a VPU (Voice Pick Unit, a voice pickup unit). ) bone conduction sensor.

As shown in FIG. 1 , the headset device may include: a processor 1001 , such as a CPU, a communication bus 1002 , a user interface 1003 , a network interface 1004 , and a memory 1005 . Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. Optionally, the network interface 1004 may include a standard wired interface and a wireless interface (eg, a Wi-Fi interface). The memory 1005 may be high-speed RAM memory, or may be non-volatile memory, such as disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .

Those skilled in the art can understand that the structure of the earphone device shown in FIG. 1 does not constitute a limitation on the earphone device, and may include more or less components than the one shown, or combine some components, or arrange different components.

As shown in FIG. 1 , the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a call noise reduction program based on a headset device.

In the terminal shown in FIG. 1 , the network interface 1004 is mainly used to connect to the background server and perform data communication with the background server; the user interface 1003 is mainly used to connect to the client and perform data communication with the client; and the processor 1001 can be used for The call noise reduction program based on the headset device stored in the memory 1005 is called, and the following operations are performed:

Determine whether the wearer of the earphone device is outputting voice information according to the digital sound signal detected by the bone conduction sensor;

If so, then determine whether the wearer is wearing the mask correctly according to the analog signal obtained by the analog microphone collecting the voice information;

When it is determined that the wearer is wearing the mask correctly, and it is determined that the wearer makes a voice call based on the headset device, the first voice signal collected by the bone conduction sensor and the second voice collected by the analog microphone are used. Signal for call noise reduction.

Optionally, the processor 1001 can also be configured to call the headset device-based call noise reduction program stored in the memory 1005, and perform the following operations:

extracting a low-frequency voice signal from the first voice signal collected by the bone conduction sensor when the wearer makes a voice call;

extracting a high-frequency voice signal from the second voice signal collected by the analog microphone when the wearer makes a voice call;

The low-frequency voice signal and the high-frequency voice signal are mixed into one channel of voice data output for noise reduction of calls.

Optionally, the processor 1001 can also be configured to call the headset device-based call noise reduction program stored in the memory 1005, and perform the following operations:

acquiring the digital sound signal detected by the bone conduction sensor, and performing speech recognition on the digital sound signal;

When voice recognition is performed on the digital sound signal to detect voice information from the digital sound signal, it is determined that the wearer of the earphone device is outputting voice information.

Optionally, the processor 1001 can also be configured to call the headset device-based call noise reduction program stored in the memory 1005, and perform the following operations:

acquiring an analog signal obtained by collecting the voice information continuously output by the wearer by the analog microphone;

Whether the wearer is wearing the mask correctly is determined according to the proportional relationship between the high-frequency energy and the low-frequency energy in the analog signal.

Optionally, the processor 1001 can also be configured to call the headset device-based call noise reduction program stored in the memory 1005, and perform the following operations:

After performing fast Fourier transform on the analog signal, calculate the high-frequency energy and the low-frequency energy in the analog signal respectively, and determine the ratio relationship between the high-frequency energy and the low-frequency energy;

If the proportion relationship is that the proportion of the low-frequency energy exceeds a preset threshold, it is determined that the wearer is wearing a mask correctly;

If the proportion relationship is that the proportion of the low-frequency energy exceeds a preset threshold, it is determined that the wearer does not wear a mask or does not wear a mask correctly.

Optionally, the processor 1001 can also be configured to call the headset device-based call noise reduction program stored in the memory 1005, and determine whether the wearer is After the steps to properly wear a mask, do the following:

When it is determined that the wearer does not wear a mask or does not wear a mask correctly, a prompt voice for reminding to wear a mask properly is output through the speaker of the earphone device.

Optionally, the headset device establishes a Bluetooth connection with the mobile terminal in advance, and the processor 1001 can also be used to call the headset device-based call noise reduction program stored in the memory 1005, and perform the following operations:

When receiving the voice Bluetooth signal transmitted by the mobile terminal through the Bluetooth connection, it is determined that the wearer is currently making a voice call based on the headset device.

Based on the above-mentioned earphone device, various embodiments of the method for noise reduction of a call based on the earphone device of the present invention are proposed. In each embodiment of the motion monitoring method for a wireless headset of the present invention, the call noise reduction method based on the headset device of the present invention is applied to the above-mentioned headset device configured with an analog microphone and a bone conduction sensor.

Please refer to FIG. 2 , which is a schematic flowchart of a first embodiment of a method for noise reduction of calls based on a headset device of the present invention. In the first embodiment of the call noise reduction method based on the headset device of the present invention, the call noise reduction method based on the headset device of the present invention includes:

Step S10: determining whether the wearer of the earphone device is outputting voice information according to the digital sound signal detected by the bone conduction sensor;

In this embodiment, the earphone device continuously collects the data sound signal generated by the wearer through the bone conduction sensor configured by itself during the whole process of detecting and determining that it is worn, and then determines the current Whether the wearer is outputting voice information at the moment.

It should be noted that, in this embodiment, since the digital sound signal collected by the VPU bone conduction sensor is transmitted from the skull, that is, the transmission channel of the VPU bone conduction sensor and the ordinary microphone to collect air medium sound waves are different, and the VPU bone conduction sensor has different transmission channels. The sound waves collected by the conductive sensor are hardly affected by wearing a mask and are more suitable for speech recognition detection operations than ordinary microphones. Based on this, the headset device uses the VPU bone conduction sensor to assist the MIC analog microphone to perform signal processing to identify the wearer's wearing state of the mask and to reduce the noise of the mask friction interference during the wearer's call.

Further, in some feasible embodiments, step S10 may include:

Step S101, acquiring a digital sound signal detected by the bone conduction sensor, and performing speech recognition on the digital sound signal;

Step S102, when voice recognition is performed on the digital sound signal to detect voice information from the digital sound signal, it is determined that the wearer of the earphone device is outputting voice information.

In this embodiment, when the earphone device continuously collects the digital sound signal generated by the wearer of the earphone device through the bone conduction sensor configured by itself, it first obtains the digital sound signal collected by the bone conduction sensor in real time, and then uses The algorithm module performs speech recognition on the digital sound signal. In this way, when the earphone device recognizes the voice information sent by the wearer from the digital sound signal by performing voice recognition on the digital sound signal, it is immediately determined that the wearer is outputting the voice information at the current moment.

In addition, in some feasible embodiments, when the headphone device performs speech recognition on the digital signal collected by the bone conduction sensor, if the headphone device does not recognize the voice information from the digital sound signal, the headphone device will determine The wearer is not currently outputting voice information.

Step S20, if yes, then determine whether the wearer is wearing the mask correctly according to the analog signal obtained by collecting the voice information with the analog microphone;

In this embodiment, after determining that the wearer of the headset device is outputting voice information at the current moment based on the digital sound signal collected by the bone conduction sensor, the headset device further conducts the voice information output by the wearer according to the analog microphone configured by itself. The obtained analog signal is collected to determine whether the wearer is wearing the mask correctly at the current moment.

Further, in some feasible embodiments, step S20 may include:

Step S201, acquiring an analog signal obtained by collecting the voice information continuously output by the wearer by the analog microphone;

In this embodiment, after determining that the wearer of the headset device is outputting voice information at the current moment according to the digital sound signal collected by the bone conduction sensor, the headset device immediately obtains the voice information continuously output by the analog microphone configured by itself for the wearer Acquire the analog signal.

Step S202: Determine whether the wearer is wearing the mask correctly according to the ratio of the high-frequency energy and the low-frequency energy in the analog signal.

In this embodiment, immediately after acquiring the analog signal obtained by the analog microphone that collects the voice information continuously output by the wearer, the earphone device performs fast Fourier transform on the analog signal into the frequency domain, and then calculates the analog signal respectively. The high-frequency energy and low-frequency energy in the signal are calculated, and the proportional relationship between the high-frequency energy and the low-frequency energy is calculated, so as to determine whether the wearer is wearing the mask correctly at the current moment based on the proportional relationship.

It should be noted that, in this embodiment, the earphone device calculates high-frequency energy (greater than 4KHZ) and low-frequency energy with the frequency of 4KHZ as the threshold after the fast Fourier transform of the model signal collected by the analog microphone into the frequency domain. (less than 4KHZ).

Further, in some feasible embodiments, the above-mentioned step S202 may specifically include:

Step S2021, after performing fast Fourier transform on the analog signal, calculate the high-frequency energy and the low-frequency energy in the analog signal respectively, and determine the ratio relationship between the high-frequency energy and the low-frequency energy;

Step S2022, if the proportion relationship is that the proportion of the low-frequency energy exceeds a preset threshold, it is determined that the wearer is wearing a mask correctly;

Step S2022, if the proportion relationship is that the proportion of the low-frequency energy exceeds a preset threshold, it is determined that the wearer does not wear a mask or does not wear a mask correctly.

In this embodiment, the earphone device performs fast Fourier transform on the analog signal collected by the analog microphone into the frequency domain, and calculates the high-frequency energy and low-frequency energy in the analog signal respectively, and calculates the high-frequency energy and the low-frequency energy. After the ratio between the low-frequency energies, if the ratio calculated by the headphone device is that the ratio of the low-frequency energy exceeds the preset threshold, the headphone device immediately determines that the wearer of the headphone device is wearing a mask correctly at the current moment, or , if the proportion relationship calculated by the earphone device is that the proportion of low frequency energy does not exceed the preset threshold, the earphone device immediately determines that the wearer of the earphone device does not wear a mask correctly or does not wear a mask at the current moment.

It should be noted that, in this embodiment, the above-mentioned preset threshold may specifically be that the earphone device preliminarily collects the analog signal of the wearer wearing the mask correctly through the analog microphone, and simulates the collected analog signal through an algorithm. When the determined wearer wears the mask correctly, the minimum proportion of low-frequency energy in the analog signal corresponding to the voice information output by the wearer. It should be understood that, based on different practical applications, the types of earphone devices themselves or the algorithms used by them may be different. Thus, the specific size of the preset threshold determined by the earphone devices based on the above process can of course be different. The present invention is based on The call noise reduction method of the headset device is not limited to the specific size of the preset threshold.

In addition, in some feasible embodiments, after the step of “determining whether the wearer is wearing the mask correctly according to the analog signal obtained by collecting the voice information by the analog microphone” in the above step S20, the present invention is based on the The call noise reduction method may also include:

Step A, when it is determined that the wearer does not wear a mask or does not wear a mask correctly, output a prompt voice for reminding to wear a mask properly through the speaker of the earphone device.

In this embodiment, the ratio between the high-frequency energy and the low-frequency energy in the calculated analog signal by the earphone device is that the proportion of low-frequency energy does not exceed the preset threshold, so it is determined that the wearer of the earphone device has no energy at the current moment. When wearing a mask correctly or not wearing a mask, the headset device further outputs a prompt voice to remind the wearer to wear the mask properly through the speaker configured by itself.

Exemplarily, as shown in the application process shown in FIG. 3 , the earphone device (assumed to be a TWS earphone) performs AD conversion (analog-to-digital conversion) on the analog signal collected by the wearer when outputting the voice information through the analog microphone configured by itself, so as to convert the The analog signal is converted into a digital signal, and then the TWS headset transmits the digital signal and the digital sound signal generated by the wearer collected by the VPU bone conduction sensor into its own algorithm module;

After that, the TWS headset uses the digital sound signal collected by the VPU bone conduction sensor to detect the VAD voice recognition through the algorithm module. voice information is recognized), then the TWS headset further uses the same frame of the digital signal corresponding to the analog signal collected by the analog microphone when the wearer outputs the voice information to perform mask wearing detection and judgment. That is, the TWS headset performs FFT (fast Fourier) conversion on the digital signal corresponding to the analog signal through its own mask detection module into the frequency domain and calculates the high-frequency energy (greater than 4KHZ) and the low-frequency energy (less than 4K) respectively, so, When the proportion of low-frequency energy of the digital signal corresponding to several frames of analog signals in which the wearer speaks continuously is high (higher than the above-mentioned preset threshold), the TWS headset will judge that the wearer is wearing the mask correctly at the current moment, otherwise TWS The headset judges that the wearer is not wearing a mask at the current moment or that the mask is incorrectly worn;

Furthermore, when the TWS headset judges that the wearer is wearing the mask correctly, the TWS headset will set the mask wearing flag to "true", and if the TWS headset judges that the wearer is not wearing a mask at the current moment or the mask is incorrectly worn, the TWS headset not only Set the mask wearing flag to "false", and at the same time play a prompt sound "please wear a mask" through the speaker to remind the wearer to wear a mask properly.

Step S30, when it is determined that the wearer is wearing the mask correctly, and when it is determined that the wearer makes a voice call based on the headset device, the first voice signal collected by the bone conduction sensor and the voice signal collected by the analog microphone are used. The second voice signal performs call noise reduction.

In this embodiment, the earphone device determines that the wearer is wearing the mask correctly at the current moment based on the ratio of high-frequency energy and low-frequency energy in the above-mentioned analog signal, and further determines that the wearer is using the earphone at the current moment. When the device conducts a voice call, the headset device will collect the first voice signal in real time during the wearer's call according to the bone conduction sensor configured by itself, and the analog microphone configured by itself during the wearer's call. The second voice signal is used to perform noise reduction for the voice signal output by the wearer during the voice call process.

It should be noted that, because the software logic in the TWS headset usually triggers the audio stream of the call and music playback through the Bluetooth signal, that is, the general audio stream is only available in the call and music scene after the Bluetooth connection. Therefore, on the one hand, the TWS headset needs to open a new audio stream after confirming that it is worn by the user, which is specially used for the process of detecting whether the mask is correctly worn as described in the above steps S10 and S20. On the other hand, the TWS headset determines whether the user is currently making a voice call based on itself based on whether the Bluetooth signal transmitted by the connected terminal device is received.

Based on this, in a feasible embodiment, the above-mentioned headset device establishes a Bluetooth connection with the mobile terminal in advance, and the method for noise reduction of a call based on the headset device of the present invention may further include:

Step B, when receiving the voice Bluetooth signal transmitted by the mobile terminal through the Bluetooth connection, determine that the wearer is currently making a voice call based on the headset device.

In this embodiment, the headset device establishes a Bluetooth connection with a smart terminal device (such as a smart phone, a tablet computer, a smart TV, a personal computer and other terminal devices) through a Bluetooth module configured by itself, so that the headset device can be based on the Bluetooth connection. Receive the bluetooth data transmitted by the smart terminal device and output it to the wearer of the headset device to form an audio stream or a voice stream of a call internally. In this way, when the headset device receives the voice Bluetooth signal transmitted by the smart terminal device through the Bluetooth connection, it immediately initializes the audio data that it is outputting, and determines that the wearer of the headset device is conducting communication with other devices based on the headset device at the current moment. Voice calls between users of smart terminal devices.

Further, in a feasible embodiment, in the above-mentioned step S30, the step of "reducing noise during a call according to the first voice signal collected by the bone conduction sensor and the second voice signal collected by the analog microphone". , which can include:

Step S301, extracting a low-frequency voice signal from a first voice signal collected by the bone conduction sensor when the wearer conducts a voice call;

Step S302, extracting a high-frequency voice signal from the second voice signal collected by the analog microphone when the wearer conducts a voice call;

Step S302, mixing the low-frequency voice signal and the high-frequency voice signal into one channel of voice data output for noise reduction of the call.

In this embodiment, the earphone device performs noise reduction for the voice signal output by the wearer according to the first voice signal and the second voice signal collected in real time by the bone conduction sensor and the analog microphone respectively during the wearer's call process When the headset device first extracts the low-frequency voice signal from the first voice signal, and synchronously or asynchronously extracts the high-frequency voice signal from the second voice signal, then the headset device combines the low-frequency voice signal with the high-frequency voice signal. The voice signals are mixed to form 1 channel of voice data, and the voice data is output to the smart terminal device through the Bluetooth connection with the smart terminal device for noise reduction of the call.

It should be noted that, in this embodiment, after the TWS headset triggers the call audio process through Bluetooth, if the TWS headset determines that the above-mentioned mask wearing flag is set to "true" at this time, the TWS headset will be in the normal call data processing process. Add an algorithm for noise reduction of mask noise. Because the interference caused by the friction of the mask is mainly concentrated in the low frequency, the digital sound signal collected by the VPU bone conduction sensor is collected by the bone conduction transmission method, that is, there is almost no interference sound caused by the friction of the mask in the low frequency of the digital sound signal. Therefore, The TWS headset will use the low frequency of the digital sound signal collected by the VPU bone conduction sensor during the wearer's voice call as the audio of the algorithm output signal.

Exemplarily, as shown in FIG. 4 , in this embodiment, the TWS earphone performs noise reduction processing on the interference sound generated by the friction of the mask as follows:

(1) The TWS headset performs FFT on the first voice signal collected by the VPU bone conduction sensor during the wearer's voice call and the second voice signal collected by the MIC analog microphone during the voice call at the same time. The operation forms the corresponding VPU frequency domain data and MIC frequency domain data;

(2) The TWS headset inputs the VPU frequency domain data into the LP low-pass filter to extract the low-frequency signal below 1KHz in the first voice signal, and inputs the MIC data into the HP high-pass filter to extract the second voice signal above 1KHz. High-frequency signal, and the TWS headset mixes the low-frequency signal and the high-frequency signal to form a complete channel of voice data;

(3) TWS tunes the 1-channel voice data formed after mixing through EQ (equalizer) to brighten the sound, and then outputs it through the Bluetooth connection with the smart terminal device.

In this way, in this embodiment, the earphone device continuously collects the data sound signal generated by the wearer through the bone conduction sensor configured by itself during the whole process of detecting and determining that it is worn, and then based on the collected digital sound The signal determines whether the wearer is outputting voice information at the current moment. After that, the earphone device further determines whether the wearer is wearing the mask correctly at the current moment according to the analog signal collected by the analog microphone configured by itself for the voice information output by the wearer. Finally, when the headset device determines that the wearer is making a voice call based on the headset device at the current moment, the headset device will, according to the bone conduction sensor configured by the headset device, collect the first voice signal in real time during the wearer's call process, and match it with its own The configured analog microphone collects the second voice signal in real time during the wearer's call to perform noise reduction for the voice signal output by the wearer during the voice call.

In this way, the embodiment of the present invention realizes that the bone conduction sensor and the simulated microphone can be used to accurately detect whether the user is wearing the mask correctly, and, in the case of detecting that the user is wearing the mask, the bone conduction sensor and the simulated microphone are further used to detect whether the user is wearing a mask correctly. Noise reduction is performed on the voice signal during the voice call using the headset device, thereby improving the voice quality of the user's call based on the headset device.

In addition, an embodiment of the present invention also provides a call noise reduction device based on a headset device. The call noise reduction device based on the headset device of the present invention is applied to a headset device configured with an analog microphone and a bone conduction sensor. Please refer to FIG. 5 . FIG. 5 is a schematic diagram of functional modules of an embodiment of the headset device-based call noise reduction device of the present invention. As shown in FIG. 5 , the headset device-based call noise reduction device of the present invention includes:

The mask detection module 10 is used to determine whether the wearer of the headset device is outputting voice information according to the digital sound signal detected by the bone conduction sensor; and, if so, collect the voice information obtained according to the analog microphone. The analog signal determines whether the wearer is wearing the mask correctly;

The call noise reduction module 20 is configured to, when it is determined that the wearer is wearing a mask correctly, and when it is determined that the wearer makes a voice call based on the earphone device, according to the first voice signal collected by the bone conduction sensor and the The second voice signal collected by the analog microphone is used for noise reduction of the call.

Optionally, the call noise reduction module 20 includes:

a low-frequency voice extraction unit, configured to extract a low-frequency voice signal from the first voice signal collected by the bone conduction sensor when the wearer conducts a voice call;

a high-frequency voice extraction unit, configured to extract a high-frequency voice signal from the second voice signal collected by the analog microphone when the wearer conducts a voice call;

The mixed output unit is used for mixing the low-frequency voice signal and the high-frequency voice signal into one channel of voice data output for noise reduction of calls.

Optionally, the mask detection module 10 includes:

a first acquiring unit, configured to acquire a digital sound signal detected by the bone conduction sensor, and perform speech recognition on the digital sound signal;

The first determining unit is configured to determine that the wearer of the earphone device is outputting voice information when voice recognition is performed on the digital voice signal to detect voice information from the digital voice signal.

Optionally, the mask detection module 10 further includes:

a second acquiring unit, configured to acquire an analog signal obtained by the analog microphone collected for the voice information continuously output by the wearer;

The second determination unit is configured to determine whether the wearer is wearing the mask correctly according to the ratio between the high-frequency energy and the low-frequency energy in the analog signal.

Optionally, the second determining unit includes:

The proportion determination subunit is used to respectively calculate the high frequency energy and the low frequency energy in the analog signal after performing the fast Fourier transform on the analog signal, and determine the difference between the high frequency energy and the low frequency energy. proportional relationship;

a first determination subunit, configured to determine that the wearer is wearing a mask correctly if the ratio of the ratio is that the ratio of the low-frequency energy exceeds a preset threshold;

The second determination subunit is configured to determine that the wearer does not wear a mask or does not wear a mask correctly if the ratio of the ratio is that the ratio of the low-frequency energy exceeds a preset threshold.

Optionally, the call noise reduction device based on the headset device of the present invention further includes:

The prompt module is configured to output a prompt voice for reminding the wearer to wear the mask properly through the speaker of the earphone device when it is determined that the wearer does not wear the mask or does not wear the mask correctly.

Optionally, the call noise reduction device based on the headset device of the present invention further includes:

The call scene determination module is configured to determine that the wearer is currently making a voice call based on the headset device when receiving the voice Bluetooth signal transmitted by the mobile terminal through the Bluetooth connection.

The specific embodiments of each functional module of the headset device-based call noise reduction device of the present invention during operation are basically the same as the above-mentioned embodiments of the headset device-based call noise reduction program method of the present invention, and will not be repeated here.

The present invention also provides a computer storage medium on which a call noise reduction program based on an earphone device is stored, and when the above-mentioned call noise reduction program based on the earphone device is executed by a processor, it is implemented as described in any of the above embodiments The steps of a call noise reduction program method based on a headset device.

The specific embodiments of the computer storage medium of the present invention are basically the same as the above-mentioned embodiments of the call noise reduction program method based on the headset device of the present invention, and are not repeated here.

The present invention also provides a computer program product, the computer program product includes a computer program, and when the computer program is executed by a processor, implements the method for noise reduction of a call based on a headset device according to any of the above embodiments of the present invention. The steps are not repeated here.

It should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or system comprising a series of elements includes not only those elements, It also includes other elements not expressly listed or inherent to such a process, method, article or system. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article or system that includes the element.

The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages or disadvantages of the embodiments.

From the description of the above embodiments, those skilled in the art can clearly understand that the method of the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is better implementation. Based on such understanding, the technical solutions of the present invention can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products are stored in a storage medium (such as ROM/RAM) as described above. , magnetic disk, optical disk), including several instructions to make a headset device (which may be a TWS headset, etc.) to execute the methods described in the various embodiments of the present invention.

The above are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied in other related technical fields , are similarly included in the scope of patent protection of the present invention.

Claims (10)

1. A conversation noise reduction method based on earphone equipment is characterized in that the conversation noise reduction method based on earphone equipment is applied to the earphone equipment configured with an analog microphone and a bone conduction sensor, and comprises the following steps:

determining whether a wearer of the earphone device is outputting voice information from the digital sound signal detected by the bone conduction sensor;

if yes, determining whether the wearer correctly wears the mask according to a simulation signal obtained by collecting the voice information by the simulation microphone;

when the fact that the wearer correctly wears the mask is determined, and the fact that the wearer carries out voice call based on the earphone device is determined, call noise reduction is carried out according to a first voice signal collected by the bone conduction sensor and a second voice signal collected by the analog microphone.

2. The method of claim 1, wherein the step of reducing the noise of the call according to the first voice signal collected by the bone conduction sensor and the second voice signal collected by the analog microphone comprises:

extracting a low-frequency voice signal from a first voice signal acquired by the bone conduction sensor when the wearer carries out voice communication;

extracting a high-frequency voice signal from a second voice signal acquired by the analog microphone when the wearer carries out voice call;

and mixing the low-frequency voice signal and the high-frequency voice signal into 1-path voice data to be output so as to reduce the noise of the call.

3. The headset-device-based call noise reduction method of claim 1, wherein the step of determining whether the wearer of the headset device is outputting voice information from the digital sound signal detected by the bone conduction sensor comprises:

acquiring a digital sound signal detected by the bone conduction sensor, and performing voice recognition on the digital sound signal;

determining that a wearer of the headset device is outputting voice information upon voice recognition of the digital sound signal to detect voice information from the digital sound signal.

4. The method for reducing noise of a call based on an earphone device as claimed in claim 1, wherein the step of determining whether the wearer is wearing the mask correctly according to the analog signal obtained by collecting the voice information by the analog microphone comprises:

acquiring an analog signal acquired by the analog microphone aiming at the voice information continuously output by the wearer;

and determining whether the wearer correctly wears the mask according to the ratio relation between the high-frequency energy and the low-frequency energy in the analog signal.

5. The method for reducing noise of a call based on an earphone device as claimed in claim 4, wherein the step of determining whether the wearer is wearing the mask correctly according to the ratio between the high frequency energy and the low frequency energy in the analog signal comprises:

after carrying out fast Fourier transform on the analog signal, respectively calculating high-frequency energy and low-frequency energy in the analog signal, and determining the ratio relationship between the high-frequency energy and the low-frequency energy;

if the proportion relation is that the low-frequency energy proportion exceeds a preset threshold value, determining that the wearer correctly wears the mask;

and if the proportion relation is that the low-frequency energy proportion exceeds a preset threshold value, determining that the wearer does not wear the mask or does not wear the mask correctly.

6. The headset-device-based call noise reduction method according to any one of claims 1 to 5, wherein after the step of determining whether the wearer is wearing a mask correctly according to the analog signal obtained by collecting the voice information by the analog microphone, the method further comprises:

and when the wearer is determined not to wear the mask or not to wear the mask correctly, outputting a prompt voice for reminding the wearer of wearing the mask according to the standard through a loudspeaker of the earphone device.

7. The method for reducing the noise of the conversation based on the earphone device according to any one of the claims 1 to 5, wherein the earphone device establishes a Bluetooth connection with the mobile terminal in advance, and the method further comprises the following steps:

and when receiving a voice Bluetooth signal transmitted by the mobile terminal through the Bluetooth connection, determining that the wearer is currently carrying out voice call based on the earphone equipment.

8. The utility model provides a conversation based on earphone equipment falls device of making an uproar which characterized in that, conversation based on earphone equipment falls the device of making an uproar and is applied to the earphone equipment that disposes analog microphone and bone conduction sensor, conversation based on earphone equipment falls the device of making an uproar and includes:

the mask detection module is used for determining whether a wearer of the earphone device outputs voice information according to the digital sound signals detected by the bone conduction sensor; if yes, determining whether the wearer correctly wears the mask according to an analog signal obtained by acquiring the voice information by the analog microphone;

and the conversation noise reduction module is used for determining that the wearer correctly wears the mask and determining that the wearer can make a conversation noise reduction according to the first voice signal acquired by the bone conduction sensor and the second voice signal acquired by the analog microphone when making a voice conversation based on the earphone equipment.

9. An earphone device, characterized in that the earphone device comprises: a memory, a processor and a headset device based call noise reduction program stored on the memory and executable on the processor, the headset device based call noise reduction program when executed by the processor implementing the steps of the headset device based call noise reduction method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a headset device based call noise reduction program, which when executed by a processor implements the steps of the headset device based call noise reduction method according to any one of claims 1 to 7.

Images