Headphone audio system and control method based on array sensor module

Abstract

The present invention discloses an earphone audio system and control method based on an array sensor module, and relates to the field of earphone audio processing technology, including a noise reduction demand entry module, an environmental audio monitoring unit, an environmental sound analysis unit, and an array audio sensor unit. The noise reduction demand entry module of the present invention provides a more specific noise reduction demand entry method, and the environmental noise data monitored interactively by the microprocessor and the environmental sound analysis unit are displayed on a touch screen in combination with the current direction data, providing users with more accurate noise reduction demand options; after the user selects the noise reduction demand of the corresponding direction by operating the touch screen or recording the voice through the microphone of the earphone, the array audio sensor module transmits the reference reflection signal after adaptive filtering and delay adjustment, and the reference reflection signal should have the same frequency and amplitude as the noise, but the opposite phase, so that the specified direction can be targeted and directional and accurate noise reduction can be performed, and the noise reduction effect is outstanding.

Description

Technical Field

[0001] The present invention relates to the technical field of earphone audio processing, and in particular to an earphone audio system and a control method based on an array sensor module. Background Art

[0002] The current noise reduction technology for headphones is a technology that reduces or eliminates ambient noise by physical or electronic means, which is divided into passive noise reduction and active noise reduction.

[0003] Passive noise reduction technology mainly uses physical methods to block noise, such as using sound insulation materials, sealed earmuff designs, etc., to reduce external noise from entering the ears.

[0004] Earmuffs use high-density foam or other sound-isolating materials to absorb and block noise. The better the seal of the earmuff design, the better the passive noise reduction effect. The defects of physical noise reduction technology are:

[0005] Low-frequency noise is generally not well isolated. To improve noise reduction, the earmuffs may need to fit more tightly, which may reduce comfort for long-term wear. Passive noise reduction cannot completely eliminate all noise, especially low-frequency noise. In very noisy environments, passive noise reduction is not ideal.

[0006] Active noise reduction technology uses a built-in microphone to capture external noise, and then the built-in digital signal processor (DSP) analyzes the frequency and waveform of the noise and generates sound waves with opposite phases to cancel the noise. At present, this noise reduction technology cannot achieve targeted directional noise reduction due to the relatively single sound unit setting, and the noise reduction effect is not ideal. Summary of the invention

[0007] In order to solve the above-mentioned technical problem of headphone noise reduction, the present invention provides a headphone audio system and control method based on an array sensor module. The following technical solutions are adopted:

[0008] An earphone audio system based on an array sensor module, comprising a noise reduction demand recording module, an ambient audio monitoring unit, an ambient sound analysis unit and an array audio sensor unit, wherein the noise reduction demand recording module is installed on the earphone, and the user's active noise reduction demand is recorded based on voice or keystrokes, the ambient audio monitoring unit is communicatively connected to the noise reduction demand recording module, and is used to monitor the ambient audio around the earphone, the ambient sound analysis unit is communicatively connected to the ambient audio monitoring unit, and the ambient sound analysis unit determines the main frequency components of the noise based on the ambient audio, and establishes a mathematical model of the noise signal, including amplitude, frequency and phase;

[0009] The data input end of the array audio sensor module communicates with the noise reduction demand input module and the ambient sound analysis unit to exchange the user's active noise reduction demand data and the mathematical model data of the noise signal, and generates a reference reflection signal according to the active noise reduction demand data and the noise model. The reference reflection signal has the same frequency and amplitude as the noise, but the opposite phase;

[0010] The array audio sensor module transmits the reference reflection signal after adaptive filtering and delay adjustment, and the array audio sensor unit and the environmental audio monitoring unit interact with the environmental audio after the reference reflection signal is transmitted.

[0011] Optionally, the noise reduction demand entry module includes a first magnetic field sensor, a microprocessor and a touch display, which are integrated on the outer shell of the headset or the wire control, and the first magnetic field sensor is used to monitor the current direction. The microprocessor is respectively communicated with the first magnetic field sensor, the touch display and the ambient sound analysis unit to control the touch display to display the current direction information and display the noise reduction degree selection items based on the direction information.

[0012] Optionally, the noise reduction demand entry module also includes a voice analysis chip, which communicates with the microphone of the headset to interact with the user's voice active noise reduction demand, and the voice analysis chip communicates with the microprocessor to interact with voice analysis result data.

[0013] By adopting the above technical solution, the noise reduction demand input module can provide a more specific way to input noise reduction demand. Specifically, a touch screen is used to provide an input medium, which can be installed on the line control of the headset. The user holds the touch screen flat, and the first magnetic field sensor can collect the current direction data of the touch screen. The microprocessor processes the direction data and controls the touch screen to display the current direction data. When the touch screen is flat and lit, the ambient audio monitoring unit and the ambient sound analysis unit are both turned on. At this time, the microprocessor can interact with the ambient sound analysis unit to monitor the ambient noise data, and combine it with the current direction data on the touch screen, and then display the noise reduction degree selection items to provide users with more accurate noise reduction demand options;

[0014] After the user selects the noise reduction requirement in the corresponding direction by operating the touch display screen or recording voice through the microphone of the headset, the array audio sensor module will transmit the reference reflection signal after adaptive filtering and delay adjustment. The reference reflection signal should have the same frequency and amplitude as the noise, but the opposite phase. In this way, targeted and precise noise reduction can be performed in the specified direction, and the noise reduction effect is outstanding.

[0015] Optionally, the environmental audio monitoring unit includes multiple miniature environmental sound microphones and a second magnetic field sensor. The multiple miniature environmental sound microphones are evenly distributed in multiple microphone mounting holes provided on the outer shell of the earphone. The second magnetic field sensor is installed on the outer shell of the earphone for monitoring the current orientation of the earphone. The multiple miniature environmental sound microphones and the second magnetic field sensor are respectively communicated with the environmental sound analysis unit.

[0016] By adopting the above technical solution, the environmental audio monitoring unit adopts an array-type receiving unit layout, which is installed using the earphone casing, and a second magnetic field sensor is installed at the same time to achieve more accurate direction information collection.

[0017] Optionally, the ambient sound analysis unit includes an ambient audio analysis chip and a first memory, wherein the first memory is respectively communicatively connected to a plurality of miniature ambient sound receiving microphones and a second magnetic field sensor, the first memory stores a common ambient audio feature database and a corresponding noise signal mathematical model, the ambient audio analysis chip is communicatively connected to the memory, an audio feature matching algorithm is used to identify ambient audio features, and the corresponding ambient audio features are located through a sound source localization algorithm, the stored noise signal mathematical model is matched based on the ambient audio features and the positioning results, and the recognition results, positioning results and noise signal mathematical model are stored in the first memory.

[0018] By adopting the above technical solution, the functions of the environmental audio analysis chip are: first, analyzing the environmental audio and positioning information collected by multiple miniature environmental sound receiving microphones and the second magnetic field sensor; second, using an audio feature matching algorithm to identify environmental audio features, and locating the corresponding environmental audio features through a sound source localization algorithm, and matching the stored noise signal mathematical model based on the environmental audio features and the positioning results;

[0019] The specific audio feature matching algorithm adopts the following steps: pre-processing the collected audio signal, such as denoising, enhancement, normalization, etc., to improve the accuracy and efficiency of subsequent processing.

[0020] Audio features are extracted from the preprocessed audio signal. These features may include but are not limited to Mel-frequency cepstral coefficients (MFCCs), spectral centroid, spectral flatness, pitch, rhythm, etc.

[0021] The purpose of feature extraction is to convert audio signals into feature vectors that can be used for comparison and classification.

[0022] The extracted audio features are matched with different environmental audio features pre-stored in the first memory. The matching algorithm is such as K-NN, SVM, NN, etc.

[0023] The algorithm is used to calculate the similarity or distance between the current audio feature and the stored feature. The similarity or distance can be calculated by a variety of methods, such as Euclidean distance, cosine similarity, etc. According to the threshold of similarity or distance, the environment type that best matches the current audio data is determined. If the similarity is higher than a certain threshold or the distance is lower than a certain threshold, it is considered that a matching environmental audio feature has been found.

[0024] The mathematical model of noise signal is established by following the steps below:

[0025] There are different mathematical models for different types of noise. Common noise types include:

[0026] White noise: Noise with equal power at all frequencies.

[0027] Pink noise: power is inversely proportional to frequency.

[0028] Brownian noise (red noise): The power is inversely proportional to the square of the frequency.

[0029] Impulse Noise: Randomly occurring short-duration pulses of energy.

[0030] Analyze the statistical characteristics of noise, including:

[0031] Mean: The average value of the noise.

[0032] Variance: The volatility of the noise.

[0033] Probability density function (PDF): describes the probability of the noise taking different values.

[0034] Autocorrelation function: describes the correlation of noise at different time points.

[0035] According to the statistical characteristics of noise, the following mathematical model can be established:

[0036] For white noise:

[0037] Probability density function: Assume a Gaussian distribution (normal distribution).

[0038] For ideal white noise, the autocorrelation function is the variance when the lag is zero and zero otherwise.

[0039] For pink noise:

[0040] Power Spectral Density: Power is inversely proportional to frequency.

[0041] For impulse noise: Probability density function: Assumed to be a mixed Gaussian distribution, containing a Gaussian distribution representing a normal signal and a Gaussian distribution representing a pulse.

[0042] Use statistical methods (such as maximum likelihood estimation, least squares method, etc.) to estimate model parameters.

[0043] The accuracy of the model is verified by experimental data. The following methods are used:

[0044] Goodness-of-fit test: such as the chi-square test, checks how well the model fits the actual data.

[0045] Cross-validation: Use one part of the data to build a model and another part of the data to validate the model.

[0046] The established model is applied to noise suppression.

[0047] Optionally, the array audio sensor unit includes a buffer, a reflected sound controller based on a control chip, a power amplifier and multiple sound units, the buffer communicates with the first memory to exchange recognition results, positioning results and a mathematical model of the noise signal, the reflected sound controller is communicatively connected to the buffer, generates control instructions for respectively controlling the multiple sound units based on the interactive recognition results, positioning results and the mathematical model of the noise signal, and controls the emission of reflected sound waves of the multiple sound units respectively through the power amplifier.

[0048] By adopting the above technical solution, the reflected sound controller is a control chip, which generates a suppression signal based on the inverse of the mathematical model of the noise signal, and generates a control signal based on the noise model and control strategy of the suppression signal. The control signal is a signal with an opposite phase to the noise, which is used to acoustically cancel the noise. The power amplifier can realize power adjustment of multiple sound units, thereby achieving selective noise reduction.

[0049] Optionally, the sound unit includes a horn-shaped mounting piece, a base and a directional sound speaker. A plurality of sound unit mounting holes are provided on the outer shell of the earphone. The horn-shaped mounting piece is installed at the sound unit mounting hole with the opening facing outward. The base is installed on the inner bottom of the horn-shaped mounting piece. The directional sound speaker is detachably mounted on the base. The directional sound speaker is communicatively connected to the signal output end of the power amplifier.

[0050] Optionally, the horn-shaped mounting and the base are molded from a sound-isolating material.

[0051] By adopting the above technical scheme, the sound insulation material can be a sound insulation board, a sound absorbing board, etc. The horn-shaped mounting part and the base formed by the sound insulation material mold can basically isolate the reverse interference of the directional sound speaker to the sound field of the earphone itself. The audio signal emitted by the directional sound speaker is directed outward to suppress the environmental noise.

[0052] The earphone audio control method based on the array sensor module adopts the earphone audio system based on the array sensor module to respond to the environmental noise reduction demand and comprises the following steps:

[0053] Step 1: The user holds the touch screen horizontally in front of the body, turns on the first magnetic field sensor, monitors the current direction of the touch screen, and operates the touch screen to select voice or key to record the user's active noise reduction needs;

[0054] Step 2: The ambient audio monitoring unit starts to collect ambient audio data around the headset, and communicates with the ambient sound analysis unit to exchange ambient audio data;

[0055] Step 3: The ambient audio analysis chip uses an audio feature matching algorithm to identify ambient audio features, and locates the corresponding ambient audio features through a sound source localization algorithm;

[0056] Step 4, the microprocessor controls the touch display screen to display the current direction information, and displays the current noise intensity, the corresponding direction and the noise reduction degree adjustment button based on the eight direction information; the eight direction information are the south direction, the southeast direction, the east direction, the northeast direction, the north direction, the northwest direction, the west direction and the southwest direction;

[0057] The current noise intensity is the average value of the regional noise in the corresponding direction;

[0058] Step 5: The user operates the touch screen to select the corresponding direction and noise reduction level;

[0059] Step 6: The environmental audio analysis chip uses an audio feature matching algorithm to identify environmental audio features, and locates the corresponding environmental audio features through a sound source localization algorithm, and matches the stored noise signal mathematical model based on the environmental audio features and the localization results;

[0060] Step 7: The reflected sound controller generates control instructions for controlling multiple sound units based on the interactive recognition results, positioning results and mathematical model of the noise signal, and controls multiple sound units to emit reflected sound waves in the selected direction through the power amplifier to achieve active noise reduction in the specified direction.

[0061] Optionally, in step 6, a KNN nearest neighbor matching algorithm is used to identify the environmental audio features; and an SVM sound source localization algorithm is used to locate the corresponding environmental audio features.

[0062] In summary, the present invention includes at least one of the following beneficial technical effects:

[0063] The present invention can provide an earphone audio system and control method based on an array sensor module. The noise reduction requirement input module provides a more specific noise reduction requirement input method. The microprocessor and the ambient sound analysis unit interactively monitor the ambient noise data, and display it on the touch screen in combination with the current direction data, and then display the noise reduction degree selection items, so as to provide users with more accurate noise reduction requirement selection items.

[0064] After the user selects the noise reduction requirement in the corresponding direction by operating the touch display screen or recording voice through the microphone of the headset, the array audio sensor module will transmit the reference reflection signal after adaptive filtering and delay adjustment. The reference reflection signal should have the same frequency and amplitude as the noise, but the opposite phase. In this way, targeted and precise noise reduction can be performed in the specified direction, and the noise reduction effect is outstanding. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a schematic diagram of the communication connection of components of the earphone audio system based on the array sensor module of the present invention;

[0066] Figure 2 It is a schematic diagram of the structural principle of the earphone audio system based on the array sensor module of the present invention installed on the earphone housing;

[0067] Figure 3 It is a schematic diagram of the structure of the sound unit of the earphone audio system based on the array sensor module of the present invention.

[0068] Description of the drawings: 1. Noise reduction demand input module; 11. First magnetic field sensor; 12. Microprocessor; 13. Touch screen; 14. Voice analysis chip; 2. Environmental audio monitoring unit; 21. Miniature environmental sound microphone; 22. Second magnetic field sensor; 3. Environmental sound analysis unit; 31. Environmental audio analysis chip; 32. First memory; 41. Buffer; 42. Reflection sound controller; 43. Power amplifier; 44. Multiple sound units; 441. Speaker-shaped mounting piece; 442. Base; 443. Directional sound speaker; 100. Headphones. DETAILED DESCRIPTION

[0069] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0070] The embodiment of the present invention discloses an earphone audio system and a control method based on an array sensor module.

[0071] Reference Figure 1-Figure 3 , Embodiment 1, an earphone audio system based on an array sensor module, comprising a noise reduction demand recording module 1, an environmental audio monitoring unit 2, an environmental sound analysis unit 3 and an array audio sensor unit, wherein the noise reduction demand recording module 1 is installed on the earphone 100, and the user's active noise reduction demand is recorded based on voice or keystrokes, the environmental audio monitoring unit 2 is communicatively connected to the noise reduction demand recording module 1, and is used to monitor the environmental audio around the earphone 100, the environmental sound analysis unit 3 is communicatively connected to the environmental audio monitoring unit 2, and the environmental sound analysis unit 3 determines the main frequency components of the noise based on the environmental audio, and establishes a mathematical model of the noise signal, including amplitude, frequency and phase;

[0072] The data input end of the array audio sensor module communicates with the noise reduction demand input module and the ambient sound analysis unit to exchange the user's active noise reduction demand data and the mathematical model data of the noise signal, and generates a reference reflection signal according to the active noise reduction demand data and the noise model. The reference reflection signal has the same frequency and amplitude as the noise, but the opposite phase;

[0073] The array audio sensor module transmits the reference reflected signal after adaptive filtering and delay adjustment, and the array audio sensor unit and the environmental audio monitoring unit 2 interact with the environmental audio after the reference reflected signal is transmitted.

[0074] Embodiment 2, the noise reduction demand entry module 1 includes a first magnetic field sensor 11, a microprocessor 12 and a touch display screen 13, the first magnetic field sensor 11, the microprocessor 12 and the touch display screen 13 are integrated and installed on the outer shell or the wire control of the headset 100, the first magnetic field sensor 11 is used to monitor the current direction, the microprocessor 12 is respectively communicated with the first magnetic field sensor 11, the touch display screen 13 and the ambient sound analysis unit 3, controls the touch display screen 13 to display the current direction information, and displays the noise reduction degree selection items based on the direction information.

[0075] In Example 3, the noise reduction demand input module 1 also includes a voice analysis chip 14, which communicates with the microphone of the headset 100 to interact with the user's voice active noise reduction demand, and communicates with the microprocessor 12 to interact with the voice analysis result data.

[0076] The noise reduction demand input module 1 can provide a more specific noise reduction demand input method, specifically using a touch screen 13 to provide an input medium, which can be installed on the line control of the headset 100. The user holds the touch screen 13 flat, and the first magnetic field sensor 11 can collect the current direction data of the touch screen 13. The microprocessor 12 processes the direction data and controls the touch screen 13 to display the current direction data. When the touch screen 13 is flat and lit, the environmental audio monitoring unit 2 and the environmental sound analysis unit 3 are both turned on. At this time, the microprocessor 12 can interact with the environmental sound analysis unit 3 to monitor the environmental noise data, and combine it with the current direction data on the touch screen 13 for display, and then cooperate with the display of the noise reduction degree option to provide the user with a more accurate noise reduction demand option;

[0077] After the user selects the noise reduction requirement in the corresponding direction by operating the touch display screen 13 or recording voice through the microphone of the headset 100, the array audio sensor module transmits the reference reflection signal after adaptive filtering and delay adjustment. The reference reflection signal should have the same frequency and amplitude as the noise, but the opposite phase. In this way, targeted and precise noise reduction can be performed in the specified direction, and the noise reduction effect is outstanding.

[0078] Embodiment 4, the environmental audio monitoring unit 2 includes multiple miniature environmental sound microphones 21 and a second magnetic field sensor 22, and the multiple miniature environmental sound microphones 21 are evenly installed in multiple microphone mounting holes provided on the outer shell of the earphone 100, and the second magnetic field sensor 22 is installed on the outer shell of the earphone 100, and is used to monitor the current orientation of the earphone 100. The multiple miniature environmental sound microphones 21 and the second magnetic field sensor 22 are respectively communicated with the environmental sound analysis unit 3.

[0079] The environmental audio monitoring unit 2 adopts an array-type sound receiving unit layout, is installed using the housing of the earphone 100, and is also installed with a second magnetic field sensor 22 to achieve more accurate direction information collection.

[0080] The first magnetic field sensor 11 and the second magnetic field sensor 22 are both three-axis magnetic field sensors.

[0081] Embodiment 5, the environmental sound analysis unit 3 includes an environmental audio analysis chip 31 and a first memory 32, the first memory 32 is respectively communicated with multiple miniature environmental sound receiving microphones 21 and the second magnetic field sensor 22, the first memory 32 stores a common environmental audio feature database and a corresponding noise signal mathematical model, the environmental audio analysis chip 31 is communicated with the memory 32, an audio feature matching algorithm is used to identify environmental audio features, and the corresponding environmental audio features are located through a sound source positioning algorithm, the stored noise signal mathematical model is matched based on the environmental audio features and the positioning results, and the recognition results, positioning results and noise signal mathematical model are stored in the first memory 32.

[0082] The functions of the environmental audio analysis chip 31 are: first, analyzing the environmental audio and positioning information collected by the multiple miniature environmental sound receiving microphones 21 and the second magnetic field sensor 22; second, using an audio feature matching algorithm to identify environmental audio features, and locating the corresponding environmental audio features through a sound source localization algorithm, and matching the stored noise signal mathematical model based on the environmental audio features and the positioning results;

[0083] The specific audio feature matching algorithm adopts the following steps: pre-processing the collected audio signal, such as denoising, enhancement, normalization, etc., to improve the accuracy and efficiency of subsequent processing.

[0084] Audio features are extracted from the preprocessed audio signal. These features may include but are not limited to Mel-frequency cepstral coefficients (MFCCs), spectral centroid, spectral flatness, pitch, rhythm, etc.

[0085] The purpose of feature extraction is to convert audio signals into feature vectors that can be used for comparison and classification.

[0086] The extracted audio features are matched with different environmental audio features pre-stored in the first memory 32. The matching algorithm is, for example, a nearest neighbor classifier (K-NN), a support vector machine (SVM), a neural network (NN), and the like.

[0087] The algorithm is used to calculate the similarity or distance between the current audio feature and the stored feature. The similarity or distance can be calculated by a variety of methods, such as Euclidean distance, cosine similarity, etc. According to the threshold of similarity or distance, the environment type that best matches the current audio data is determined. If the similarity is higher than a certain threshold or the distance is lower than a certain threshold, it is considered that a matching environmental audio feature has been found.

[0088] The mathematical model of noise signal is established by following the steps below:

[0089] There are different mathematical models for different types of noise. Common noise types include:

[0090] White noise: Noise with equal power at all frequencies.

[0091] Pink noise: power is inversely proportional to frequency.

[0092] Brownian noise (red noise): The power is inversely proportional to the square of the frequency.

[0093] Impulse Noise: Randomly occurring short-duration pulses of energy.

[0094] Analyze the statistical characteristics of noise, including:

[0095] Mean: The average value of the noise.

[0096] Variance: The volatility of the noise.

[0097] Probability density function (PDF): describes the probability of the noise taking different values.

[0098] Autocorrelation function: describes the correlation of noise at different time points.

[0099] According to the statistical characteristics of noise, the following mathematical model can be established:

[0100] For white noise:

[0101] Probability density function: Assume a Gaussian distribution (normal distribution).

[0102] For ideal white noise, the autocorrelation function is the variance when the lag is zero and zero otherwise.

[0103] For pink noise:

[0104] Power Spectral Density: Power is inversely proportional to frequency.

[0105] For impulse noise: Probability density function: Assumed to be a mixed Gaussian distribution, containing a Gaussian distribution representing a normal signal and a Gaussian distribution representing a pulse.

[0106] Use statistical methods (such as maximum likelihood estimation, least squares method, etc.) to estimate model parameters.

[0107] The accuracy of the model is verified by experimental data. The following methods are used:

[0108] Goodness-of-fit test: such as the chi-square test, checks how well the model fits the actual data.

[0109] Cross-validation: Use one part of the data to build a model and another part of the data to validate the model.

[0110] The established model is applied to noise suppression.

[0111] Embodiment 6, the array audio sensor unit includes a buffer 41, a reflection sound controller 42 based on a control chip, a power amplifier 43 and multiple sound units 44, the buffer 41 communicates with the first memory 32 to exchange recognition results, positioning results and a mathematical model of a noise signal, the reflection sound controller 42 is communicatively connected with the buffer 41, and generates control instructions for controlling the multiple sound units 44 respectively based on the interaction recognition results, positioning results and the mathematical model of the noise signal, and controls the emission of reflected sound waves of the multiple sound units 44 respectively through the power amplifier 43.

[0112] The reflected sound controller 42 is a control chip that generates a suppression signal based on the inverse of the mathematical model of the noise signal, and generates a control signal based on the noise model and control strategy of the suppression signal. The control signal is a signal with an opposite phase to the noise, which is used to acoustically cancel the noise. The power amplifier 43 can realize power adjustment of multiple sound units 44, thereby achieving selective noise reduction.

[0113] Embodiment 7, the sound unit 44 includes a horn-shaped mounting piece 441, a base 442 and a directional sound speaker 443, a plurality of sound unit mounting holes are provided on the outer shell of the earphone 100, the horn-shaped mounting piece 441 is installed at the sound unit mounting hole with the opening facing outward, the base 442 is installed on the inner bottom of the horn-shaped mounting piece 441, the directional sound speaker 443 is detachably mounted on the base 442, and the directional sound speaker 443 is communicatively connected with the signal output end of the power amplifier 43.

[0114] In Example 8, the trumpet-shaped mounting member 441 and the base 442 are both formed by a mold made of sound insulation material.

[0115] The sound insulation material can be a sound insulation board, a sound absorbing board, etc. The horn-shaped mounting part 441 and the base 442 formed by the sound insulation material mold can basically isolate the reverse interference of the directional sound speaker 443 to the sound field of the earphone 100 itself. The audio signal emitted by the directional sound speaker 443 is directed outward to suppress the environmental noise.

[0116] Embodiment 9, a headphone audio control method based on an array sensor module, uses a headphone audio system based on an array sensor module to respond to and control environmental noise reduction requirements, and includes the following steps:

[0117] Step 1: The user holds the touch screen 13 horizontally in front of the body, turns on the first magnetic field sensor 11, monitors the current direction of the touch screen 13, and operates the touch screen 13 to select voice or key to record the user's active noise reduction needs;

[0118] Step 2, the environmental audio monitoring unit 2 starts collecting environmental audio data around the earphone 100, and communicates with the environmental sound analysis unit 3 to exchange environmental audio data;

[0119] Step 3, the environmental audio analysis chip 31 uses an audio feature matching algorithm to identify environmental audio features, and locates the corresponding environmental audio features through a sound source localization algorithm;

[0120] Step 4, the microprocessor 12 controls the touch display screen 13 to display the current direction information, and displays the current noise intensity, the corresponding direction and the noise reduction degree adjustment button based on the eight direction information; the eight direction information are the south direction, the southeast direction, the east direction, the northeast direction, the north direction, the northwest direction, the west direction and the southwest direction;

[0121] The current noise intensity is the average value of the regional noise in the corresponding direction;

[0122] Step 5, the user operates the touch display screen 13 to select the corresponding direction and noise reduction level;

[0123] Step 6: The environmental audio analysis chip 31 uses an audio feature matching algorithm to identify environmental audio features, and locates the corresponding environmental audio features through a sound source localization algorithm, and matches the stored noise signal mathematical model based on the environmental audio features and the localization results;

[0124] Step 7, the reflected sound controller 42 generates control instructions for controlling multiple sound units 44 respectively based on the interactive recognition results, positioning results and mathematical model of the noise signal, and controls multiple sound units 44 to emit reflected sound waves in the selected direction through the power amplifier 43, thereby realizing active noise reduction in the specified direction.

[0125] In Example 10, in step 6, the KNN nearest neighbor matching algorithm is used to identify the environmental audio features; and the SVM sound source localization algorithm is used to locate the corresponding environmental audio features.

[0126] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An earphone audio system based on an array sensor module, characterized in that: The invention comprises a noise reduction demand recording module (1), an ambient audio monitoring unit (2), an ambient sound analysis unit (3) and an array audio sensor unit, wherein the noise reduction demand recording module (1) is installed on the earphone (100) and records the user's active noise reduction demand based on voice or keystrokes, the ambient audio monitoring unit (2) is communicatively connected to the noise reduction demand recording module (1) and is used to monitor the ambient audio around the earphone (100), the ambient sound analysis unit (3) is communicatively connected to the ambient audio monitoring unit (2), and the ambient sound analysis unit (3) determines the main frequency components of the noise based on the ambient audio and establishes a mathematical model of the noise signal, including amplitude, frequency and phase; The data input end of the array audio sensor module communicates with the noise reduction demand input module and the ambient sound analysis unit to exchange the user's active noise reduction demand data and the mathematical model data of the noise signal, and generates a reference reflection signal according to the active noise reduction demand data and the noise model. The reference reflection signal has the same frequency and amplitude as the noise, but the opposite phase; The array audio sensor module transmits the reference reflection signal after adaptive filtering and delay adjustment, and the array audio sensor unit and the environmental audio monitoring unit (2) interact with the environmental audio after the reference reflection signal is transmitted; The noise reduction demand input module (1) comprises a first magnetic field sensor (11), a microprocessor (12) and a touch display screen (13); the first magnetic field sensor (11), the microprocessor (12) and the touch display screen (13) are integrated and mounted on an outer shell or a wire control of the headset (100); the first magnetic field sensor (11) is used to monitor the current direction; the microprocessor (12) is respectively connected to the first magnetic field sensor (11), the touch display screen (13) and the ambient sound analysis unit (3) for communication, and controls the touch display screen (13) to display the current direction information, and displays the current noise intensity, the corresponding direction and the noise reduction degree adjustment button based on the eight direction information; The eight directions are due south, southeast, due east, northeast, due north, northwest, due west and southwest.

2. The earphone audio system based on array sensor module according to claim 1, characterized in that: The noise reduction demand input module (1) also includes a voice analysis chip (14), wherein the voice analysis chip (14) communicates with the microphone of the headset (100) to exchange the user's voice active noise reduction demand, and the voice analysis chip (14) communicates with the microprocessor (12) to exchange voice analysis result data.

3. The earphone audio system based on array sensor module according to claim 2, characterized in that: The environmental audio monitoring unit (2) comprises a plurality of miniature environmental sound receiving microphones (21) and a second magnetic field sensor (22). The plurality of miniature environmental sound receiving microphones (21) are evenly installed in a plurality of microphone installation holes provided on the housing of the earphone (100). The second magnetic field sensor (22) is installed on the housing of the earphone (100) and is used to monitor the current orientation of the earphone (100). The plurality of miniature environmental sound receiving microphones (21) and the second magnetic field sensor (22) are respectively connected to the environmental sound analysis unit (3) for communication.

4. The earphone audio system based on array sensor module according to claim 3, characterized in that: The environmental sound analysis unit (3) comprises an environmental audio analysis chip (31) and a first memory (32). The first memory (32) is respectively connected to a plurality of miniature environmental sound receiving microphones (21) and a second magnetic field sensor (22). The first memory (32) stores a common environmental audio feature database and a corresponding noise signal mathematical model. The environmental audio analysis chip (31) is connected to the first memory (32) for communication, and uses an audio feature matching algorithm to identify environmental audio features, and uses a sound source positioning algorithm to locate the corresponding environmental audio features. The stored noise signal mathematical model is matched based on the environmental audio features and the positioning result, and the recognition result, the positioning result and the noise signal mathematical model are stored in the first memory (32).

5. The earphone audio system based on array sensor module according to claim 4, characterized in that: The array audio sensor unit comprises a buffer (41), a reflection sound controller (42) based on a control chip, a power amplifier (43) and a plurality of sound units (44); the buffer (41) communicates with a first memory (32) to exchange an interactive recognition result, a positioning result and a mathematical model of a noise signal; the reflection sound controller (42) is communicatively connected to the buffer (41), generates control instructions for controlling the plurality of sound units (44) based on the interactive recognition result, the positioning result and the mathematical model of the noise signal, and controls the emission of reflected sound waves by the plurality of sound units (44) through the power amplifier (43).

6. The earphone audio system based on array sensor module according to claim 5, characterized in that: The sound unit (44) comprises a horn-shaped mounting piece (441), a base (442) and a directional sound-emitting speaker (443); a plurality of sound-emitting unit mounting holes are provided on the housing of the earphone (100); the horn-shaped mounting piece (441) is mounted at the sound-emitting unit mounting hole with its opening facing outward; the base (442) is mounted on the inner bottom of the horn-shaped mounting piece (441); the directional sound-emitting speaker (443) is detachably mounted on the base (442); and the directional sound-emitting speaker (443) is communicatively connected to a signal output end of the power amplifier (43).

7. The earphone audio system based on array sensor module according to claim 6, characterized in that: The horn-shaped mounting piece (441) and the base (442) are both formed by a sound insulation material mold.

8. An earphone audio control method based on an array sensor module, characterized in that: The earphone audio system based on the array sensor module of claim 7 is used to respond to and control the environmental noise reduction requirements, comprising the following steps: Step 1: The user holds the touch display screen (13) horizontally in front of the user, turns on the first magnetic field sensor (11), monitors the current direction of the touch display screen (13), and operates the touch display screen (13) to select voice or keystrokes to input the user's active noise reduction needs; Step 2, the environmental audio monitoring unit (2) starts collecting environmental audio data around the earphone (100), and communicates with the environmental audio analysis unit (3) to exchange the environmental audio data; Step 3, the ambient audio analysis chip (31) uses an audio feature matching algorithm to identify ambient audio features, and locates the corresponding ambient audio features through a sound source localization algorithm; Step 4, the microprocessor (12) controls the touch display screen (13) to display current direction information, and displays the current noise intensity, the corresponding direction and the noise reduction degree adjustment button based on eight direction information; the eight direction information are due south, southeast, due east, northeast, due north, northwest, due west and southwest; The current noise intensity is the average value of the regional noise in the corresponding direction; Step 5: The user operates the touch display screen (13) to select a corresponding direction and noise reduction level; Step 6, the ambient audio analysis chip (31) uses an audio feature matching algorithm to identify ambient audio features, and locates the corresponding ambient audio features through a sound source localization algorithm, and matches the stored noise signal mathematical model based on the ambient audio features and the localization results; Step 7: The reflected sound controller (42) generates control instructions for controlling the multiple sound generating units (44) based on the interactive recognition result, the positioning result and the mathematical model of the noise signal, and controls the multiple sound generating units (44) to emit reflected sound waves in the selected direction through the power amplifier (43), thereby realizing active noise reduction in the specified direction.

9. The earphone audio control method based on array sensor module according to claim 8, characterized in that: In step 6, the KNN nearest neighbor matching algorithm is used to identify the environmental audio features; and the SVM sound source localization algorithm is used to locate the corresponding environmental audio features.

Images