Volume equalization type Bluetooth sound equipment facing on-site environment sound and light collection

Abstract

This invention discloses a volume-equalizing Bluetooth speaker for ambient sound and light acquisition. It features a separate cavity for the feedforward microphone and speaker, with a Z-shaped sound guide tube and internal sound-absorbing components added between the feedforward microphone and the pickup hole. This utilizes the forced change in sound wave propagation direction due to the bent path and physical attenuation of mid-to-high frequencies, allowing the feedforward microphone to primarily receive low-frequency ambient acoustic signals, thus improving the signal-to-noise ratio of the acquired ambient acoustic signals. Furthermore, the ambient acoustic signal is decomposed into multi-band energy values ​​and their cross-frame variation rates. These are combined with the illumination variation rate to comprehensively determine the scene and output a scene mode with physical representation. The corresponding mode retrieves preset volume equalization parameters, and the total sound pressure level and total harmonic distortion within the cavity are extracted based on the cavity sound pressure signal acquired by the feedback microphone, serving as a closed-loop monitoring basis for volume adjustment. This achieves scene-adaptive volume equalization control while ensuring speaker safety.

Description

Technical Field

[0001] This invention relates to the field of audio technology, and more specifically, to a volume-equalized Bluetooth speaker for ambient sound and light acquisition. Background Technology

[0002] Bluetooth speakers refer to audio devices with built-in Bluetooth chips that use Bluetooth connections instead of traditional wired connections. As users' demands for audio experience continue to increase, Bluetooth speakers have been widely used in various environments such as cars and outdoors. Traditional Bluetooth speakers generally use manual adjustment for volume control, which cannot perceive real-time changes in ambient acoustics and lighting conditions, resulting in a mismatch between volume output and the ambient environment, which seriously affects the listening experience.

[0003] Currently, some audio equipment manufacturers are attempting to incorporate environmental sensing for intelligent volume compensation. For example, the smart speaker and its sound effect control method disclosed in patent number CN119521054A introduces a light sensor to switch sound effect modes based on ambient light brightness. However, it relies solely on a single optical sensor and completely ignores the dynamic changes in environmental acoustic characteristics, making it unable to perceive the impact of acoustic environments such as road noise, wind noise, and rain noise on the listening experience.

[0004] For ambient noise acquisition, conventional audio equipment such as speakers and headphones often use built-in microphones. These microphones are typically mounted directly on the speaker housing or on a circuit board inside the housing, using pickup holes in the housing to capture external ambient acoustic signals. However, this conventional microphone mounting method has inherent acquisition limitations when the audio is playing:

[0005] On the one hand, the sound played by the speaker is easily transmitted to the environmental acquisition microphone through the air cavity inside the housing and the solid surface of the housing wall. This results in the signal picked up by the microphone containing a large number of speaker playback sound components, making it difficult for the system to accurately distinguish between "real environmental noise" and "speaker playback sound crosstalk".

[0006] On the other hand, in on-site environments such as vehicles or outdoors, wind noise generated by airflow directly impacting the microphone diaphragm and solid-conducted vibration noise generated by the vibration of the housing by the speaker can seriously pollute the acquired signal. Compensating for volume based on the polluted environmental acoustic signal can easily lead to volume misjudgment.

[0007] Therefore, in response to the actual shortcomings, a volume-equalizing Bluetooth speaker designed for on-site sound and light acquisition is proposed. Summary of the Invention

[0008] The purpose of this invention is to solve the above-mentioned problems, and a volume equalization Bluetooth speaker for on-site environmental sound and light acquisition is provided.

[0009] The objective of this invention can be achieved through the following technical solution: a volume-equalizing Bluetooth speaker for ambient sound and light acquisition, comprising a decorative cover, a sound-emitting enclosure installed inside the rear end of the decorative cover, a speaker and a feedback microphone arranged adjacent to the speaker diaphragm installed inside the sound-emitting enclosure, a position-adaptive feedforward microphone and a Z-shaped sound guide tube installed inside the front end of the decorative cover, a pickup hole corresponding to the inlet end of the Z-shaped sound guide tube opened on the edge of the upper wall of the decorative cover, and an ambient light sensor also embedded in the upper wall of the decorative cover;

[0010] The decorative cover houses a volume equalization control panel integrating a digital signal processor and a power amplifier. The digital signal processor includes:

[0011] The on-site environmental information acquisition module is electrically connected to the feedback microphone, the feedforward microphone, and the ambient light sensor, respectively, and is used to synchronously acquire environmental acoustic signals, real-time sound pressure signals, and real-time ambient illuminance values.

[0012] The acoustic feature extraction module is used to perform frequency band energy analysis on the environmental acoustic signal to output multi-frequency band environmental energy values ​​to the scene collaborative judgment module, and to perform sound pressure level and distortion analysis on the real-time sound pressure signal to output intracavity sound pressure feature values ​​containing total sound pressure level and total harmonic distortion to the volume equalization drive module.

[0013] The optical feature extraction module is used to calculate the first derivative of the real-time ambient illuminance value with respect to time and output the illuminance change rate to the scene collaborative judgment module.

[0014] The scene collaborative determination module is used to identify scenes based on the coupling relationship between multi-frequency band environmental energy values ​​and illuminance change rates, and output scene patterns with physical representations.

[0015] The volume equalization driver module is electrically connected to the scene coordination judgment module and the power amplifier. It is used to retrieve the corresponding preset volume equalization parameters according to the scene mode, and to correct and limit the output of the power amplifier based on the intracavity sound pressure characteristic value.

[0016] Furthermore, an isolation plate is installed inside the decorative cover, which divides the interior of the decorative cover into a rear sound-emitting cavity and a front sound-absorbing cavity. The sound-emitting box is installed in the rear sound-emitting cavity through an elastic damping element, and a sound-emitting cover plate that connects to the outlet of the sound-emitting box is configured at the rear end of the rear sound-emitting cavity. The feedforward microphone is fixedly installed at the bottom of the front sound-absorbing cavity through a PCB bracket, and its diaphragm is set upward.

[0017] Furthermore, the Z-shaped sound guide tube is composed of a first vertical section, a horizontal section, and a second vertical section connected in sequence. The outlet end of the second vertical section corresponds to the diaphragm of the feedforward microphone, and an air gap of 3mm-6mm is reserved between the two.

[0018] Furthermore, the inner wall of the Z-shaped sound guide tube is covered with sound-absorbing components at the bends and horizontal sections. The sound-absorbing components are corrugated porous polyurethane sound-absorbing sponges.

[0019] Furthermore, the specific process of performing frequency band energy analysis on environmental acoustic signals is as follows:

[0020] The acoustic feature extraction module performs frame processing on the environmental acoustic signal using a preset time-domain window, and performs a fast Fourier transform on each frame to convert the time-domain signal of the environmental acoustic signal into a frequency-domain spectrum. The frequency-domain spectrum is then integrated in preset low-frequency bands, preset mid-frequency bands, and preset high-low frequency bands to output the corresponding low-frequency sub-band energy values, mid-frequency sub-band energy values, and high-frequency sub-band energy values. The first-order difference between the energy values ​​of each sub-band between adjacent time windows is calculated as the cross-frame change rate. The energy values ​​of each sub-band and the corresponding cross-frame change rate are then unified as the multi-frequency band environmental energy value.

[0021] Furthermore, the specific method for performing sound pressure level and distortion analysis on real-time sound pressure signals is as follows:

[0022] The acoustic feature extraction module calculates the root mean square value of the real-time sound pressure signal in a preset time window to obtain the root mean square voltage value. Based on the voltage-sound pressure sensitivity coefficient preset by the feedback microphone, the root mean square voltage value is converted into the corresponding effective sound pressure value. Then, logarithmic operation is performed to obtain the total sound pressure level in the cavity in decibels.

[0023] A fast Fourier transform is performed on the real-time sound pressure signal to extract the energy values ​​of its fundamental frequency component and each harmonic component. The total harmonic distortion is calculated according to the definition formula of the total harmonic distortion. The specific content of the definition formula of the total harmonic distortion is: the ratio of the square root of the sum of the squares of the energy of all harmonic components to the fundamental frequency energy is multiplied by 100% to obtain the total harmonic distortion.

[0024] The total sound pressure level and total harmonic distortion within the cavity are unified as the characteristic value of the sound pressure level within the cavity.

[0025] Furthermore, the specific process by which the optical feature extraction module obtains the rate of change of illuminance is as follows:

[0026] Within a preset time window, a sliding window average filter is applied to the real-time ambient illuminance value, and the difference between the current illuminance value and the previous illuminance value after filtering is divided by the sampling interval to obtain the illuminance change rate.

[0027] Furthermore, the specific content of the scene mode determination by the scene collaboration determination module includes:

[0028] The scene collaborative determination module receives the low-frequency sub-band energy value, mid-frequency sub-band energy value, high-frequency sub-band energy value and their respective cross-frame change rates, and also receives the illumination change rate. It uses these parameter values ​​as scene determination values, compares each scene determination value with its corresponding determination threshold, and performs scene determination according to preset determination combination rules, outputting a scene mode with physical representation.

[0029] The scene modes include a first scene mode that represents drastic changes in illumination and stable high-frequency sound field, a second scene mode that represents stable illumination and sudden increase in high-frequency sound energy, a third scene mode that represents synchronous changes in sound energy across the entire frequency band, a fourth scene mode that represents periodic energy mutations in the mid-frequency range, and a fifth scene mode that represents a normal, stable scene.

[0030] Furthermore, the specific working process of the volume equalization driver module is as follows:

[0031] The volume equalization driver module has a built-in scene-equalization parameter mapping table. The mapping table has preset volume equalization parameters for each frequency band that correspond one-to-one with all the above scene modes. When any scene mode is received, the preset volume equalization parameters for each frequency band corresponding to the scene-equalization parameter mapping table are retrieved as the basic gain parameters.

[0032] The total sound pressure level is received in real time, and the loudness compensation coefficient required for each frequency band under the current total sound pressure level is calculated according to the preset equal loudness compensation curve. The compensation coefficient is superimposed on the basic gain parameter to generate the corrected gain parameter.

[0033] Simultaneously, the total harmonic distortion (THD) is received in real time. The system determines whether the THD exceeds a preset distortion safety threshold. If not, the corrected gain parameter is used as the final gain parameter and output to the power amplifier. Otherwise, based on the corrected gain parameter, the final gain parameter output to the power amplifier is gradually reduced by a preset step size until the THD falls below the preset distortion safety threshold.

[0034] Compared with the prior art, the advantages of this invention are:

[0035] 1. This solution involves installing an outer decorative cover and a sound-generating enclosure with unequal inner and outer sections, thus isolating the feedforward microphone from the speaker cavity. This physically cuts off the main path of speaker vibrations transmitted through air and solids directly to the feedforward microphone. Simultaneously, a Z-shaped sound guide tube is added between the pickup hole and the feedforward microphone. The bending path of the Z-shaped sound guide tube forces a change in the direction of sound wave propagation. Combined with the physical attenuation of mid-to-high frequency sound waves by the sound-absorbing components on the tube wall, the feedforward microphone mainly receives low-frequency environmental acoustic signals. This effectively attenuates the mid-to-high frequency interference components of the speaker's sound that enter the acquisition path through reflection and crosstalk. This physically improves the signal-to-noise ratio of the environmental acoustic signal acquisition, providing a reliable raw data foundation for subsequent accurate scene determination.

[0036] In addition, a feedback microphone is set on one side of the speaker diaphragm to collect the sound pressure signal in the cavity in front of the speaker diaphragm in real time as a closed-loop monitoring basis for volume adjustment, so as to realize real-time perception and active protection of the speaker output status.

[0037] 2. This solution also performs frequency band energy analysis on the environmental acoustic signals obtained by the above methods, converting the time-domain signals into energy values ​​of three sub-bands (low frequency, mid frequency, and high frequency) and their cross-frame change rates in the time dimension, forming multi-band environmental energy values. This improves the environmental acoustic perception from the coarse-grained level of total energy across the entire frequency band to the fine-grained level of frequency band distribution characteristics. Based on the coupling relationship between the multi-band environmental energy values ​​and the illuminance change rate, and according to the preset judgment combination rules, it analyzes and outputs scene modes with physical representations, realizing active scene recognition with sound and light coordination. Finally, based on the recognized scene mode, it retrieves the corresponding preset volume equalization parameters, and extracts the total sound pressure level and total harmonic distortion (THD) of the cavity based on the cavity sound pressure signal. It performs equal loudness correction based on the total sound pressure level and safety limiting based on the THD, achieving scene-adaptive fine-grained volume equalization control while ensuring speaker safety. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the external structure of the Bluetooth speaker of the present invention;

[0039] Figure 2 This is a cross-sectional view of the Bluetooth speaker of the present invention;

[0040] Figure 3 This is a schematic diagram of the internal structure of the sound-generating enclosure of the present invention;

[0041] Figure 4 This is a cross-sectional view of the Z-shaped sound guide tube of the present invention;

[0042] Figure 5 This is a system module block diagram of the Bluetooth speaker of the present invention.

[0043] Explanation of the labels in the diagram:

[0044] 1. Decorative cover; 2. Isolation plate; 3. Sound-emitting enclosure; 4. Speaker; 5. Sound-emitting cover plate; 6. Feedback microphone; 7. Feedforward microphone; 8. Sound pickup hole; 9. Z-shaped sound guide tube; 10. Sound-absorbing component; 11. Ambient light sensor. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0046] Example 1: This invention discloses a volume-equalizing Bluetooth speaker for ambient sound and light acquisition. Please refer to [link / reference]. Figure 1 - Figure 4 It includes a decorative cover 1, which is a hollow shell structure with openings at both ends. An isolation plate 2 is installed inside the decorative cover 1, which divides the interior of the decorative cover 1 into a rear sound-emitting cavity and a front sound-absorbing cavity.

[0047] A sound-emitting enclosure 3 is installed inside the rear sound cavity. The sound-emitting enclosure 3 is installed in the rear sound cavity through elastic damping elements. The elastic damping elements are four long rubber damping columns fixed between the isolation plate 2 and the sound-emitting cover plate 5. Multiple short rubber damping columns connected to the outer wall of the sound-emitting enclosure 3 are fixed on the long rubber damping columns. The rear end of the rear sound cavity is equipped with a sound-emitting cover plate 5 that connects to the outlet of the sound-emitting enclosure 3. A loudspeaker 4 is installed inside the sound-emitting enclosure 3. The sound-emitting enclosure 3 is suspended and installed in the rear sound cavity through the elastic damping elements, which can effectively attenuate the solid-borne noise of the loudspeaker 4 vibration transmitted to the decorative cover 1 through the enclosure, and prevent the vibration from being transmitted to the front sound-absorbing cavity and interfering with the acquisition accuracy of the feedforward microphone 7.

[0048] The sound-emitting enclosure 3 is also equipped with a feedback microphone 6. The feedback microphone 6 is located near the diaphragm of the speaker 4, and its pickup end faces the compressed air area in front of the diaphragm of the speaker 4. It is used to collect the sound pressure signal in the cavity in front of the diaphragm of the speaker 4 in real time, which is the data basis for subsequent calculation of the total sound pressure level and total harmonic distortion in the cavity.

[0049] Inside the front sound absorption cavity, there is a position-adaptive feedforward microphone 7 and a Z-shaped sound guide tube 9. The feedforward microphone 7 is fixedly installed at the bottom of the front sound absorption cavity by a PCB bracket, with its diaphragm facing upward. The feedforward microphone 7 is surrounded by a silicone sealing sleeve to isolate solid vibrations transmitted through the inner wall of the front sound absorption cavity, further ensuring the purity of the environmental acoustic signal acquisition.

[0050] The Z-shaped sound guide tube 9 is composed of a first vertical section, a horizontal section and a second vertical section connected in sequence. The upper wall edge of the decorative cover 1 is provided with a pickup hole 8 corresponding to the inlet end of the first vertical section. The outlet end of the second vertical section corresponds to the diaphragm of the feedforward microphone 7, and an air gap of 3mm-6mm is reserved between the two. The sound waves act on the diaphragm surface of the feedforward microphone 7 in a vertical incident manner.

[0051] In this preferred embodiment, the opening plane of the pickup hole 8 can be set at an elevation angle of 30-45° to the horizontal plane. When the audio system is installed on the vehicle's center console, the pickup hole 8 faces the inner surface of the vehicle's windshield to pick up low-frequency environmental reverberation signals that enter the cabin after being reflected by the windshield.

[0052] The inner wall of the Z-shaped sound guide tube 9 is covered with sound-absorbing components 10 at the bends and horizontal sections. The sound-absorbing components 10 are corrugated porous polyurethane sound-absorbing sponges with a thickness of 3mm-8mm. Specifically, the sound-absorbing components 10 are laid on the inner wall of the first bend (i.e., the corner connecting the first vertical section and the horizontal section), the lower surface of the inner wall of the horizontal section, and the inner wall of the second bend (i.e., the corner connecting the horizontal section and the second vertical section). The sound-absorbing components 10 only cover the above-mentioned key areas and do not cover the upper side and the left and right side walls of the inner wall of the horizontal section, thus preserving a smooth and rigid acoustic channel for the propagation of low-frequency plane waves.

[0053] The wavy surface increases the contact area between the sound-absorbing component 10 and the sound waves. The porous structure utilizes the friction effect between air particles and the pore walls to convert sound energy into heat energy, resulting in strong attenuation of sound waves with frequencies higher than 1kHz. Meanwhile, sound waves with frequencies lower than 300Hz, due to their longer wavelengths, can bypass or pass through the pore structure of the sound-absorbing component 10 without significant attenuation, thus achieving the physical frequency band filtering function.

[0054] An ambient light sensor 11 is also embedded in the upper wall of the decorative cover 1. Its photosensitive surface is horizontally upward or tilted at a small angle to the horizontal plane, facing the windshield or vehicle roof, and is used to sense the ambient light intensity.

[0055] Example 2: A volume equalization control panel is also installed inside the front sound absorption cavity. This control panel integrates a digital signal processor and a power amplifier. Please refer to [link / reference]. Figure 5 The digital signal processor includes a field environment information acquisition module, an acoustic feature extraction module, an optical feature extraction module, a scene coordination determination module, and a volume equalization driver module.

[0056] The on-site environmental information acquisition module is electrically connected to the feedback microphone 6, the feedforward microphone 7, and the ambient light sensor 11, respectively. It is connected to the output of the feedforward microphone 7 via the first audio input channel to acquire ambient acoustic signals, to the output of the feedback microphone 6 via the second audio input channel to acquire real-time sound pressure signals, and to the output of the ambient light sensor 11 via the optical input channel to acquire ambient optical signals. The ambient optical signals are sampled and converted into real-time ambient illuminance values, which are then converted into digital signals. The ambient acoustic signals and real-time sound pressure signals are sampled and converted into digital audio signals.

[0057] The acoustic feature extraction module, connected to the on-site environmental information acquisition module, is used to perform frequency band energy analysis on the environmental acoustic signal to output multi-frequency band environmental energy values ​​to the scene collaborative judgment module, and to perform sound pressure level and distortion analysis on the real-time sound pressure signal to output intracavity sound pressure feature values ​​containing total sound pressure level and total harmonic distortion to the volume equalization drive module.

[0058] The specific process of performing frequency band energy analysis on environmental acoustic signals is as follows:

[0059] The ambient acoustic signal is processed in frames using a preset time-domain window, and a fast Fourier transform is performed on each frame of the signal to convert the time-domain signal into a frequency-domain spectrum.

[0060] Energy integration is performed on the frequency domain spectrum within the preset low-frequency band (20Hz-200Hz), preset mid-frequency band (200Hz-2kHz), and preset high-low frequency band (2kHz-10kHz). Energy integration is the sum of the squares of all frequency components of the frequency domain spectrum within the specified frequency band range, and the corresponding outputs are the low-frequency sub-band energy value, mid-frequency sub-band energy value, and high-frequency sub-band energy value. The first-order difference of the energy values ​​of each sub-band between adjacent time windows is calculated as the cross-frame change rate. The energy values ​​of each sub-band and the corresponding cross-frame change rate are unified as the multi-band environmental energy value.

[0061] The specific method for performing sound pressure level and distortion analysis on real-time sound pressure signals is as follows:

[0062] The acoustic feature extraction module calculates the root mean square value of the real-time sound pressure signal in a preset time window, obtains the root mean square voltage value, and calculates the ratio between the root mean square voltage value and the voltage-sound pressure sensitivity coefficient based on the voltage-sound pressure sensitivity coefficient preset by the feedback microphone 6. The voltage-sound pressure sensitivity coefficient is an inherent parameter of the feedback microphone 6 measured during factory calibration. The effective sound pressure value is marked, and then the total sound pressure level in the cavity is obtained in decibels through logarithmic operation.

[0063] In addition, the acoustic feature extraction module performs a fast Fourier transform on the real-time sound pressure signal to extract the energy values ​​of its fundamental frequency component and each harmonic component, and calculates the total harmonic distortion according to the definition formula of the total harmonic distortion. The specific content of the definition formula of the total harmonic distortion is: the ratio of the square root of the sum of the squares of the energy of all harmonic components to the fundamental frequency energy, and then multiplying by 100%, to obtain the total harmonic distortion. The higher the total harmonic distortion, the more severe the waveform distortion of the output signal of the loudspeaker 4, and the greater the distortion.

[0064] The total sound pressure level and total harmonic distortion within the cavity are unified as the characteristic value of the sound pressure level within the cavity.

[0065] The optical feature extraction module is connected to the on-site environmental information acquisition module. It is used to receive real-time environmental illuminance values ​​and calculate their first derivative with respect to time. In this process, within a preset time domain window, a sliding window average filter is performed on the real-time environmental illuminance values. The difference between the current illuminance value and the previous illuminance value after filtering is divided by the sampling interval to obtain the illuminance change rate, which is then output to the scene collaborative judgment module.

[0066] The scene collaborative determination module is communicatively connected to both the acoustic feature extraction module and the optical feature extraction module. It is used to perform scene recognition based on the coupling relationship between multi-band environmental energy values ​​and illuminance change rates, and output scene patterns with physical representations. The specific process is as follows:

[0067] The system receives low-frequency subband energy values, mid-frequency subband energy values, and high-frequency subband energy values, along with their corresponding cross-frame change rates, and simultaneously receives the illumination change rate. These parameter values ​​are used as scene determination values. Each scene determination value is compared with its corresponding determination threshold, and scene determination is performed according to preset determination combination rules.

[0068] When the rate of change of illuminance is negative and its absolute value is greater than the first preset threshold, and the cross-frame change rate of the energy value of the high-frequency sub-band is less than the second preset threshold, the output is the first scene mode that represents the dramatic change of illumination and the stable high-frequency sound field. The dramatic change of illumination indicates that the sound may be entering a tunnel, underground passage or bridge hole, while the stable high-frequency sound field indicates that there is no sudden wind noise or external high-frequency interference.

[0069] When the absolute value of the rate of change of illuminance is less than the third preset threshold, and the cross-frame change rate of the high-frequency subband energy value is greater than the fourth preset threshold, the output is a second scene mode that represents stable illumination and a sudden increase in high-frequency sound energy. The physical meaning of this judgment combination is: the ambient illumination remains stable, indicating that there is no dramatic change in brightness; at the same time, the high-frequency sound energy increases dramatically in a short period of time, indicating that strong high-frequency noise suddenly appears, corresponding to the turbulent wind noise generated when the car window or sunroof is open when the vehicle is traveling at high speed.

[0070] When the cross-frame change rate of the low-frequency subband energy value, the mid-frequency subband energy value, and the high-frequency subband energy value are all greater than the fifth preset threshold and the change direction is consistent, the third scene mode representing the synchronous change of the full-band sound energy is output. The physical meaning of this judgment combination is: the ambient sound energy rises or falls synchronously in the full-band range, corresponding to the enhancement or reduction of rain noise (the noise generated by raindrops hitting the roof is broadband pink noise, and the energy of each frequency band changes synchronously), and the synchronous change of road noise caused by the uniform acceleration or deceleration of the vehicle. The feature of synchronous change of the three frequency bands and the consistent direction is the key indicator for identifying broadband ambient noise.

[0071] When a quasi-periodic pulse envelope appears in the mid-frequency sub-band energy value, and the peak energy value of the envelope exceeds the sixth preset threshold multiple of the low-frequency sub-band energy value, the fourth scene mode representing the mid-frequency periodic energy mutation is output. The physical meaning of this judgment combination is: there is a periodic sound source in the environment that is mainly mid-frequency, and the intensity of the sound source significantly exceeds the background low-frequency noise. The mid-frequency band is the core frequency range of human voice, and the quasi-periodic pulse envelope corresponds to the syllable rhythm of human voice conversation. Therefore, this judgment logic corresponds to the scene of conversation between people in the car.

[0072] When any of the above conditions are not met, the fifth scenario mode representing a normal stable scenario is output; the values ​​of the above preset thresholds are related to the specific use scenario, and those skilled in the art can determine the optimal thresholds through a limited number of real vehicle calibration experiments.

[0073] The scene collaboration judgment module outputs the judgment result to the volume equalization driver module in the form of scene mode labels.

[0074] The volume equalization driver module is electrically connected to the scene coordination judgment module and the power amplifier. It is used to retrieve the corresponding preset equalization parameters according to the scene mode and perform safe limiting based on total harmonic distortion in order to control the output gain of the power amplifier to the audio signal.

[0075] The volume equalization driver module has a built-in scene-equalization parameter mapping table. The mapping table has preset volume equalization parameters for each frequency band that correspond one-to-one with all the scene modes mentioned above. When any scene mode is received, the preset volume equalization parameters for each frequency band corresponding to the scene-equalization parameter mapping table are retrieved as the basic gain parameters.

[0076] The system receives the total sound pressure level in real time, calculates the loudness compensation coefficient required for each frequency band under the current total sound pressure level based on the preset equal loudness compensation curve, and superimposes the compensation coefficient onto the basic gain parameter to generate the corrected gain parameter. The equal loudness compensation curve here is the Fletcher-Munsen family of equal loudness curves known in the art. The loudness compensation coefficient corresponds to the sound pressure level-frequency gain correction value in the curve, and its physical dimension is the same as that of the preset volume equalization parameter, which is decibel.

[0077] Simultaneously, the total harmonic distortion (THD) is received in real time, and it is determined whether the THD exceeds the preset distortion safety threshold. In this embodiment, the preset distortion safety threshold is 5%. That is, when THD ≤ 5%, the corrected gain parameter is used as the final gain parameter output to the power amplifier. Conversely, when THD > 5%, a safety limit is triggered. Based on the corrected gain parameter, the final gain parameter output to the power amplifier is gradually reduced in a preset step size until the total harmonic distortion falls back below the preset distortion safety threshold.

[0078] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto; any equivalent substitutions or modifications made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and its improved concept, should be covered within the scope of protection of the present invention.

Claims

1. A volume-equalizing Bluetooth speaker for ambient sound and light acquisition, comprising a decorative cover (1), characterized in that: The decorative cover (1) has a sound-emitting box (3) installed inside its rear end. The sound-emitting box (3) has a speaker (4) installed inside its rear end and a feedback microphone (6) positioned close to the diaphragm of the speaker (4). The decorative cover (1) has a feedforward microphone (7) and a Z-shaped sound guide tube (9) installed inside its front end. The decorative cover (1) has a pickup hole (8) on the edge of its upper wall that corresponds to the entrance of the Z-shaped sound guide tube (9). An ambient light sensor (11) is also embedded in the upper wall of the decorative cover (1). The decorative cover (1) houses a volume equalization control panel integrating a digital signal processor and a power amplifier. The digital signal processor includes: The on-site environmental information acquisition module is electrically connected to the feedback microphone, the feedforward microphone, and the ambient light sensor, respectively, and is used to synchronously acquire environmental acoustic signals, real-time sound pressure signals, and real-time ambient illuminance values. The acoustic feature extraction module is used to perform frequency band energy analysis on environmental acoustic signals to output multi-frequency band environmental energy values, and to perform sound pressure level and distortion analysis on real-time sound pressure signals to output intracavity sound pressure feature values ​​including total sound pressure level and total harmonic distortion. The optical feature extraction module is used to calculate the first derivative of the real-time ambient illuminance value with respect to time and output the rate of change of illuminance. The scene collaborative determination module is used to identify scenes based on the coupling relationship between multi-frequency band environmental energy values ​​and illuminance change rates, and output scene patterns with physical representations. The volume equalization driver module is electrically connected to the scene coordination judgment module and the power amplifier. It is used to retrieve the corresponding preset volume equalization parameters according to the scene mode, and to correct and limit the output of the power amplifier based on the intracavity sound pressure characteristic value.

2. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 1, characterized in that: The decorative cover (1) is equipped with an isolation plate (2), which divides the interior of the decorative cover (1) into a rear sound cavity and a front sound absorption cavity. The sound-generating box (3) is installed in the rear sound cavity through an elastic damping element, and the rear end of the rear sound cavity is equipped with a sound-generating cover plate (5) that is connected to the outlet of the sound-generating box (3). The feedforward microphone (7) is fixedly installed at the bottom of the front sound absorption cavity through a PCB bracket, and its diaphragm is set upward.

3. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 2, characterized in that: The Z-shaped sound guide tube (9) is composed of a first vertical section, a horizontal section and a second vertical section connected in sequence. The outlet end of the second vertical section corresponds to the diaphragm of the feedforward microphone (7), and an air gap of 3mm-6mm is reserved between the two.

4. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 3, characterized in that: The inner wall of the Z-shaped sound guide tube (9) is covered with sound-absorbing components (10) at its bends and horizontal sections. The sound-absorbing components (10) are wavy porous polyurethane sound-absorbing sponges.

5. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 1, characterized in that: The specific process of performing frequency band energy analysis on environmental acoustic signals is as follows: The acoustic feature extraction module performs frame processing on the environmental acoustic signal using a preset time-domain window, and performs a fast Fourier transform on each frame signal to convert it into a frequency domain spectrum. The frequency domain spectrum is then integrated in preset low-frequency band, preset mid-frequency band, and preset high-low frequency band, respectively, and the corresponding low-frequency sub-band energy value, mid-frequency sub-band energy value, and high-frequency sub-band energy value are output. The first-order difference value of each sub-band energy value between adjacent time windows is calculated as the cross-frame change rate, and the energy values ​​of each sub-band and the corresponding cross-frame change rate are unified as the multi-frequency band environmental energy value.

6. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 5, characterized in that: The specific method for performing sound pressure level and distortion analysis on real-time sound pressure signals is as follows: The acoustic feature extraction module calculates the root mean square value of the real-time sound pressure signal in a preset time window, obtains the root mean square voltage value, and converts the root mean square voltage value into the corresponding effective sound pressure value according to the voltage-sound pressure sensitivity coefficient preset by the feedback microphone (6), and then obtains the total sound pressure level in the cavity in decibels through logarithmic operation; A fast Fourier transform is performed on the real-time sound pressure signal to extract the energy values ​​of its fundamental frequency component and each harmonic component. The total harmonic distortion is calculated according to the definition formula of the total harmonic distortion. The specific content of the definition formula of the total harmonic distortion is: the ratio of the square root of the sum of the squares of the energy of all harmonic components to the fundamental frequency energy is multiplied by 100% to obtain the total harmonic distortion. The total sound pressure level and total harmonic distortion within the cavity are unified as the characteristic value of the sound pressure level within the cavity.

7. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 6, characterized in that: The specific process by which the optical feature extraction module obtains the rate of change in illuminance is as follows: Within a preset time window, a sliding window average filter is applied to the real-time ambient illuminance value, and the difference between the current illuminance value and the previous illuminance value obtained after filtering is divided by the sampling interval to obtain the illuminance change rate.

8. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 7, characterized in that: The specific content of the scene mode determination module includes: The scene collaborative determination module receives the low-frequency sub-band energy value, mid-frequency sub-band energy value, high-frequency sub-band energy value and their respective cross-frame change rate, and also receives the illumination change rate. It uses these parameter values ​​as scene determination values, compares each scene determination value with its corresponding determination threshold, and performs scene determination according to the preset determination combination rules, and outputs a scene mode with physical representation. The scene modes include a first scene mode that represents drastic changes in illumination and stable high-frequency sound field, a second scene mode that represents stable illumination and sudden increase in high-frequency sound energy, a third scene mode that represents synchronous changes in sound energy across the entire frequency band, a fourth scene mode that represents periodic energy mutations in the mid-frequency range, and a fifth scene mode that represents a normal, stable scene.

9. The volume-equalizing Bluetooth speaker for ambient sound and light acquisition according to claim 8, characterized in that: The specific working process of the volume equalization driver module is as follows: The volume equalization driver module has a built-in scene-equalization parameter mapping table. The mapping table has preset volume equalization parameters for each frequency band that correspond one-to-one with all the above scene modes. When any scene mode is received, the preset volume equalization parameters for each frequency band corresponding to the scene-equalization parameter mapping table are retrieved as the basic gain parameters. The total sound pressure level is received in real time. The loudness compensation coefficient required for each frequency band under the current total sound pressure level is calculated according to the preset equal loudness compensation curve. The compensation coefficient is superimposed on the basic gain parameter to generate the corrected gain parameter. Simultaneously, the total harmonic distortion (THD) is received in real time. The system determines whether the THD exceeds a preset distortion safety threshold. If not, the corrected gain parameter is used as the final gain parameter and output to the power amplifier. Otherwise, based on the corrected gain parameter, the final gain parameter output to the power amplifier is gradually reduced by a preset step size until the THD falls below the preset distortion safety threshold.

Images