Speaker handheld status recognition and equalizer adaptive switching method, device, and equipment
By identifying the gravity fluctuation characteristics and auxiliary information of audio equipment, and dynamically adjusting the equalizer parameters, the problem of inconsistent sound quality when the audio equipment is held and placed is solved, achieving a smooth transition between states and a continuous auditory experience.
Patent Information
- Application Number
- CN202611419801.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-14
- Publication Date
- 2026-11-10
AI Technical Summary
Existing audio equipment cannot automatically adapt equalizer parameters when held or placed, resulting in inconsistent sound quality. Furthermore, existing gravity sensors have difficulty accurately identifying subtle hand-holding states.
By collecting gravity sensor signals from audio equipment, filtering and extracting wave characteristics, and combining auxiliary information to identify whether the device is held or placed, a smooth transition is achieved when the state changes, and the equalizer parameters are dynamically adjusted.
This achieves consistent sound quality across different audio equipment states, reduces the probability of misjudgment, and improves the accuracy of state recognition and the continuity of the auditory experience.
Smart Images

Figure CN123075337A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of audio signal processing technology, and in particular to a method, apparatus, and device for recognizing the handheld status of an audio device and adaptively switching the equalizer. Background Technology
[0002] Portable speakers, smart speakers, mobile display devices, and smart terminals with built-in speakers are typically equipped with one or more fixed equalizer parameters at the factory. These fixed equalizer parameters are generally calibrated according to the far-field or semi-far-field listening environment when the device is placed on a stable surface such as a desktop, stand, or wall.
[0003] In actual use, users may pick up the audio equipment from the table and hold it in their hands, or bring the audio equipment close to their body or ears and play sound while moving. At this time, the listening distance is shortened, the speaker's sound hole, low-frequency passive radiating structure, or cabinet opening may be blocked by the palm, and the coupling relationship between the equipment shell and the human body will also change, which in turn will change the low-frequency quantity, mid-high frequency directivity, near-field sound pressure, loudness perception, and dynamic headroom.
[0004] Current audio equipment typically processes audio based on manually selected sound effect modes, volume levels, playback content types, or fixed acoustic calibration parameters, lacking a mechanism to automatically switch equalizer parameters based on whether the device is held by the user. Even with built-in gravity sensors or accelerometers, existing applications are mostly used for screen orientation recognition, drop detection, or motion state judgment, primarily focusing on posture direction or large amplitude movements, and struggling to distinguish between stable placement and relatively stationary handheld states. Therefore, a solution is needed that utilizes subtle fluctuations in gravity signals combined with auxiliary information for state recognition and can smoothly switch equalizer parameters. Summary of the Invention
[0005] This disclosure aims to solve the problems that audio equipment cannot automatically adapt between far-field playback and near-field playback when placed, and that stationary handheld use is easily misjudged as a placed state, and provides a method, device, and equipment for audio equipment handheld state recognition and equalizer adaptive switching.
[0006] The first aspect of this disclosure provides a method for identifying the handheld state of an audio device and adaptively switching the equalizer, comprising: acquiring a gravity sensor signal from the audio device; filtering the gravity sensor signal and extracting fluctuation features within a preset sliding time window; determining whether the audio device is in a placed state, a handheld state, or a transitional state based on the fluctuation features; selecting a target equalizer parameter between equalizer parameters for the placed state and equalizer parameters for the handheld state according to the determined state; processing the audio signal according to the target equalizer parameter during the audio signal processing of the audio device; and smoothly transitioning the equalizer parameters when the state switches from the placed state to the handheld state or vice versa. Thus, it can automatically adapt to the far-field and near-field acoustic environments according to the actual usage state of the device, improving the consistency of sound perception under different states.
[0007] In the handheld state recognition and equalizer adaptive switching method for audio equipment disclosed in the first aspect of this disclosure, optionally, the gravity sensor signal includes X-axis data, Y-axis data, and Z-axis data output from a triaxial gravity sensor, a triaxial accelerometer, or an inertial measurement unit. Thus, the data required for state recognition can be obtained using existing sensors in the portable audio equipment.
[0008] In the audio equipment handheld state recognition and equalizer adaptive switching method disclosed in the first aspect of this disclosure, optionally, the fluctuation characteristics include one or more of the following: peak-to-peak value, variance, standard deviation, root mean square change, short-time energy, rate of change of direction, difference between adjacent sliding time windows, low-frequency jitter component, and frequency band energy. Thus, the difference between stable placement and handheld perturbation can be described from the dimensions of time domain and direction change.
[0009] In the audio equipment handheld state recognition and equalizer adaptive switching method according to the first aspect of this disclosure, optionally, determining that the audio equipment is in a handheld state includes: increasing the handheld state confidence when the fluctuation feature is higher than a handheld threshold for multiple consecutive sliding time windows; and outputting the handheld state when the handheld state confidence reaches a handheld confirmation threshold. This reduces misjudgments caused by anomalies in a single sampling point.
[0010] In the audio device handheld state recognition and equalizer adaptive switching method according to the first aspect of this disclosure, optionally, determining that the audio device is in a placed state includes: outputting the placed state when the fluctuation characteristics are below a placed threshold and continue for more than a placed confirmation time, and no touch activity, button activity, gripping event, or proximity to a human body event is detected. This allows for a reliable restoration of the placed state after the device is placed back on a stable surface.
[0011] In the audio equipment handheld state recognition and equalizer adaptive switching method disclosed in the first aspect of this disclosure, optionally, determining that the audio equipment is in a handheld state is also based on one or more auxiliary information selected from touch activity, button activity, proximity events, gripping events, posture perturbation conforming to handheld shaking mode, playback volume level, playback content type, and historical state information. Thus, it is possible to use multimodal information to assist in verifying gravity fluctuation characteristics.
[0012] In the audio equipment handheld state recognition and equalizer adaptive switching method according to the first aspect of this disclosure, optionally, when determining that the audio equipment is in a placed state, a handheld state, or a transitional state, different placement thresholds and handheld thresholds are used to form a hysteresis determination, and frequent state switching caused by sensor noise is suppressed by continuous state confirmation or state holding time. This improves the stability of the state output.
[0013] In the audio device handheld state recognition and equalizer adaptive switching method disclosed in the first aspect of this disclosure, optionally, the handheld state equalizer parameters include one or more of the following: low-frequency compensation parameters, low-frequency limiting parameters, vocal clarity parameters, and near-field comfort parameters. The corresponding handheld state equalizer parameters are selected based on the holding posture, the likelihood of speaker vent obstruction, the playback volume level, the type of playback content, or microphone feedback. This allows for audio compensation or limitation in response to near-field listening, hand occlusion, and sound pressure level variations.
[0014] In the audio device handheld state recognition and equalizer adaptive switching method disclosed in the first aspect of this disclosure, optionally, the smooth transition includes: interpolating the equalizer parameters for the placement state and the equalizer parameters for the handheld state based on the confidence level of the handheld state or the confidence level of the placement state within a preset transition time to obtain the current equalizer parameters; the preset transition time is 50 milliseconds to 1000 milliseconds. This reduces abrupt changes in timbre and loudness when picking up or putting down the device.
[0015] This disclosure provides a device for identifying the handheld state of an audio device and adaptively switching the equalizer, including a gravity sensor acquisition module, a fluctuation feature extraction module, a state determination module, an audio parameter control module, an audio signal processing module, and a smooth switching module. The gravity sensor acquisition module acquires gravity sensor signals from the audio device. The fluctuation feature extraction module filters the gravity sensor signals and extracts fluctuation features within a preset sliding time window. The state determination module determines whether the audio device is in a placed state, a handheld state, or a transitional state based on the fluctuation features. The audio parameter control module selects a target equalizer parameter between equalizer parameters for the placed state and equalizer parameters for the handheld state based on the determined state. The audio signal processing module processes the audio signal according to the target equalizer parameter during the audio signal processing of the audio device. The smooth switching module smoothly transitions the equalizer parameters when the state changes from a placed state to a handheld state or vice versa. Thus, the aforementioned state identification and adaptive equalizer switching process can be executed through a modular hardware and software architecture.
[0016] In the audio equipment handheld state recognition and equalizer adaptive switching device disclosed in the second aspect of this disclosure, the device may optionally further include an auxiliary information acquisition module. The auxiliary information acquisition module is used to acquire one or more auxiliary information from touch activity, button activity, proximity events, gripping events, playback volume level, playback content type, or historical state information. Furthermore, the state determination module determines whether the audio equipment is in a placed state, a handheld state, or a transitional state based on the auxiliary information. Therefore, the accuracy of state determination in complex interactive scenarios can be improved through multimodal auxiliary information.
[0017] In the audio equipment handheld state recognition and equalizer adaptive switching device according to the second aspect of this disclosure, optionally, the state determination module includes a confidence accumulation unit and a hysteresis determination unit. The confidence accumulation unit is used to update the handheld state confidence and the placement state confidence based on the fluctuation characteristics of a continuous sliding time window. The hysteresis determination unit is used to output the state according to different handheld thresholds and placement thresholds. This reduces the probability of a stationary handheld state being misjudged as a placement state and suppresses state jitter.
[0018] A third aspect of this disclosure provides an electronic device including a processor, a memory, a gravity sensor, a speaker, and a program stored in the memory. When the program is executed by the processor, it causes the electronic device to perform the method described in the first aspect. This enables automatic state adaptation in portable audio devices, smart speakers, mobile display devices, or smart terminals with speakers.
[0019] A fourth aspect of this disclosure provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the method described in the first aspect. This enables the deployment of state recognition and equalizer adaptive switching functions in the form of a software carrier.
[0020] According to this disclosure, the device can automatically adapt to the far-field and near-field acoustic environments based on the actual usage status of the device, thereby improving the consistency of sound perception under different conditions. Attached Figure Description
[0021] Figure 1 This is a system structure diagram illustrating the audio equipment handheld status recognition and equalizer adaptive switching device involved in the example of this disclosure.
[0022] Figure 2 This is a flowchart illustrating the placement state, handheld state, and transition state determination process involved in this disclosure example.
[0023] Figure 3 This is a schematic diagram illustrating the placement state equalizer parameters, handheld state equalizer parameters, and transition interpolation control involved in the examples of this disclosure.
[0024] Figure 4 This is a timing diagram illustrating the switching of an audio device involved in the example of this disclosure from a placed state to a handheld state and back to a placed state.
[0025] Figure labeling: 101, Gravity sensor acquisition module; 102, Auxiliary information acquisition module; 103, Wave feature extraction module; 104, Placement / Handheld state determination module (state determination module); 105, Equalizer parameter library; 106, Audio parameter control module; 107, Audio signal processing module; 108, Speaker output module. Detailed Implementation
[0026] The preferred embodiments of this disclosure are described in detail below with reference to the accompanying drawings. The drawings are schematic diagrams, and the proportions of the components do not represent actual dimensions. Unless otherwise stated, "comprising" means including the listed components but not excluding other components.
[0027] The first aspect of this disclosure provides a method for recognizing the handheld status of an audio device and adaptively switching the equalizer. This method may also be referred to as a method for recognizing the audio device's status and adapting acoustic parameters based on gravity fluctuation characteristics, a method for detecting the handheld status of a portable audio device and smoothly adjusting near-field and far-field equalization, etc., and is hereinafter simply referred to as a switching method or approach. Similarly, the second aspect of this disclosure provides a device for recognizing the handheld status of an audio device and adaptively switching the equalizer. This device may also be referred to as a device for recognizing the audio device's status and adapting acoustic parameters based on gravity fluctuation characteristics, a device for detecting the handheld status of a portable audio device and smoothly adjusting near-field and far-field equalization, etc.
[0028] like Figure 1 As shown, the audio equipment handheld status recognition and equalizer adaptive switching device may include a gravity sensor acquisition module 101, an auxiliary information acquisition module 102, a wave feature extraction module 103, a placement / handheld status determination module 104 (i.e., status determination module 104), an equalizer parameter library 105, an audio parameter control module 106, an audio signal processing module 107, and a speaker output module 108.
[0029] In the system connection relationship, the gravity sensor acquisition module 101 is connected to the wave feature extraction module 103. The auxiliary information acquisition module 102 is connected to the placement / handheld state determination module 104. The wave feature extraction module 103 is also connected to the placement / handheld state determination module 104. The placement / handheld state determination module 104 is connected to the equalizer parameter library 105 and the audio parameter control module 106, respectively, and is used to request parameter calls from the equalizer parameter library 105 and output the state determination result to the audio parameter control module 106. The audio parameter control module 106 is connected to the audio signal processing module 107 and is used to transmit the target equalizer parameters to the audio signal processing module 107. The audio signal processing module 107 is connected to the speaker output module 108 and is used to output the processed audio signal to the speaker output module 108. The audio signal processing module 107 is also connected to the equalizer parameter library 105 and is used for dynamic parameter interaction based on the audio playback state. The smooth switching module can be integrated into the audio parameter control module 106 or set as an independent logic unit to smoothly transition the equalizer parameters when the state changes.
[0030] The gravity sensor acquisition module 101 is used to acquire gravity sensor signals from the audio equipment. The fluctuation feature extraction module 103 is used to filter the gravity sensor signal and extract fluctuation features within a preset sliding time window. The placement / handheld state determination module 104 is used to determine whether the audio equipment is in a placement state, a handheld state, or a transition state based on the fluctuation features. The audio parameter control module 106 is used to select a target equalizer parameter between the placement state equalizer parameter and the handheld state equalizer parameter according to the determined state. The smooth transition module is used to smoothly transition the equalizer parameters when the state changes from placement state to handheld state or vice versa. The audio signal processing module 107 is used to process the audio signal according to the target equalizer parameter and output it to the speaker output module 108 during the audio signal processing of the audio equipment. Thus, the above-mentioned state recognition and equalizer switching process can be performed through a modular structure.
[0031] The auxiliary information acquisition module 102 is used to acquire one or more auxiliary information from touch activity, button activity, proximity events, grip events, playback volume level, playback content type, or historical status information. When performing multimodal auxiliary determination, the placement / holding state determination module 104 combines the aforementioned auxiliary information acquired by the auxiliary information acquisition module 102 to output the placement state, holding state, or transition state.
[0032] Here, the auxiliary information parameters collected by the auxiliary information acquisition module 102 are described in detail: 1. Touch activity: refers to the touch level signal generated by the user's finger touching, tapping, sliding, or long-pressing the touch sensing area or capacitive touch key on the outer shell of the audio equipment, indicating the presence of human hand contact; 2. Button activity: refers to the level change signal of the mechanical button or tactile micro switch on the audio equipment being pressed or released, indicating that the user is directly operating the equipment; 3. Proximity event: refers to the trigger event when the distance between the surface of the equipment and human skin or clothing detected by the proximity sensor is less than a preset proximity distance threshold (e.g., less than 5 cm); 4. Grip event: refers to the event when the capacitance value changes suddenly or the shell strain changes exceed the grip threshold detected by the grip detection electrodes distributed on the side or bottom of the equipment, indicating that the equipment is being hugged or squeezed by a hand; 5. Playback volume level: refers to the digital gain level or volume percentage value set by the current audio system (e.g., 0 to 100 levels), used to distinguish between near-field low sound pressure listening and far-field high-volume playback; 6. Playback content type: refers to the metadata tags of the currently playing audio (e.g., vocals, podcast, music, movie, call) or the audio content type identified through frequency domain feature classification; 7. Historical state information: refers to the previous stable state of the audio device before the current time window (e.g., whether it was continuously placed or held) and the duration of that state. All of the above auxiliary information parameters have clear sensor detection methods, physical meanings, and quantitative judgment standards.
[0033] In some examples, the gravity sensor acquisition module 101 may include a triaxial gravity sensor, a triaxial accelerometer, or an inertial measurement unit. The gravity sensor acquisition module 101 can periodically read X-axis, Y-axis, and Z-axis data. The sampling frequency of the gravity sensor signal can be from 20 Hz to 200 Hz (e.g., 50 Hz or 100 Hz). The sampling frequency represents the frequency at which the triaxial gravity sensor signal is converted from analog to digital and read per second. Within this sampling frequency range, the system can capture human micro-tremors with high temporal resolution while maintaining low system power consumption, and the parameter definitions and value ranges are clearly defined.
[0034] In some examples, determining whether an audio device is in a handheld state is also based on one or more auxiliary information, including touch activity, button activity, proximity events, gripping events, posture perturbations matching handheld shaking patterns, playback volume level, playback content type, and historical state information.
[0035] The "postural perturbation conforming to handheld shaking mode" refers to the gravity sensor signal exhibiting periodic, weak oscillations in the frequency domain, for example, between 1 Hz and 3 Hz, with the angle between the three-axis gravity vectors showing a continuous, smooth, small-amplitude drift. This mode characterizes the physiological micro-vibrations produced when a human hand naturally holds an audio device in mid-air, distinguishing it from mechanical vibrations or large-amplitude vibrations on moving vehicles. All of the aforementioned auxiliary information parameters possess measurable physical quantities and judgment criteria.
[0036] In some examples, the auxiliary information acquisition module 102 may include one or more of the following: a touch detection unit, a button detection unit, a proximity sensor, a grip detection unit, a volume level detection unit, and a playback content analysis unit.
[0037] The specific implementations of each detection unit are described below: 1. Touch detection unit: A capacitive touch sensing chip (e.g., a touch sensing IC mounted on the top cover of the device) can be used to output touch activity signals by detecting changes in the parasitic capacitance of the electrode plates; 2. Button detection unit: A general purpose input / output port (GPIO) combined with a mechanical tactile micro-switch circuit can be used to capture button press and release events by detecting high and low level transitions; 3. Proximity sensor: An infrared distance sensor, optical time-of-flight (ToF) sensor, or ultrasonic ranging sensor can be used, mounted on the side of the audio equipment casing to detect the relative distance between the human body and the casing in real time; 4. Grip detection unit: One or more pairs of capacitive sensing copper foil electrodes arranged on the inner side of the audio equipment casing (utilizing the capacitance step caused by the dielectric constant of the human body when gripping), or a piezoresistive / piezoelectric micro-strain sensor attached to the inner wall of the device casing; 5. Volume level detection unit: The current volume level can be obtained by reading the volume register value in the current audio digital signal processor (DSP) or audio codec (CODEC) through a microcontroller; 6. Playback content analysis unit: This can be implemented by an audio feature analysis algorithm run by a DSP. It can extract the spectral centroid and zero-crossing rate by performing a Fast Fourier Transform (FFT) on the current audio stream, or directly read the metadata tags of the streaming media to identify the audio type.
[0038] The auxiliary information is used to perform multimodal auxiliary verification of the fluctuation characteristics of gravity sensor signals in specific scenarios, thereby improving the robustness of state recognition.
[0039] like Figure 2As shown, the method for recognizing the handheld status of audio equipment and adaptively switching the equalizer may include steps S101 to S112. The specific implementation of each step is described in detail below with reference to the accompanying drawings.
[0040] In step S101, the gravity sensor signal of the audio equipment is acquired. Specifically, the gravity sensor acquisition module 101 periodically reads the X-axis, Y-axis, and Z-axis data output by the triaxial gravity sensor at a preset sampling frequency (e.g., 50 Hz or 100 Hz). The triaxial gravity data reflects in real time the static gravitational acceleration component and dynamic micro-motion acceleration component of the audio equipment in three-dimensional space.
[0041] In step S102, the gravity sensor signal is filtered within a preset sliding time window. After filtering, DC removal processing can be performed. The duration of the preset sliding time window can be from 200 milliseconds to 2000 milliseconds (e.g., 500 milliseconds or 1000 milliseconds), and the sliding step size can be from 50 milliseconds to 200 milliseconds (e.g., 100 milliseconds). The preset sliding time window represents the time domain span of the feature statistics. DC removal filtering eliminates the influence of the static gravity direction component on the dynamic fluctuation statistics, thereby extracting the pure dynamic fluctuation signal.
[0042] In step S103, the fluctuation characteristics of the gravity sensor signal are extracted within a preset sliding time window. The fluctuation characteristics may include one or more of the following: peak-to-peak value, variance, standard deviation, root mean square change, short-time energy, rate of change of direction, difference between adjacent sliding time windows, low-frequency jitter component, and frequency band energy.
[0043] Here, the physical meaning and calculation method of each extracted fluctuation characteristic parameter are explained: Peak-to-peak value represents the difference between the maximum and minimum values of the triaxial acceleration composite vector within the window, characterizing the range of attitude changes; variance and standard deviation represent the degree of dispersion of each sampling point within the window from the mean, characterizing the statistical dispersion of weak tremors; root mean square change represents the fluctuation amplitude of the signal's effective value between adjacent time slices; short-time energy represents the sum of the squares of the amplitudes of each sampling point after DC removal, characterizing micro-motion energy; rate of change of direction represents the angular velocity of the change of the direction angle of the gravity space vector per unit time; difference between adjacent sliding time windows represents the absolute difference between the statistical means of two consecutive windows; low-frequency jitter component represents, for example, the component corresponding to human physiological muscle tremors and respiratory movements extracted after passing through a 0.5 Hz to 5 Hz bandpass filter; frequency band energy represents, for example, the spectral integral energy within the 0.5 Hz to 5 Hz frequency band. The above fluctuation characteristic parameters have clear mathematical definitions and time-domain / frequency-domain extraction steps, and the testing and calculation methods are clear and explicit.
[0044] In some instances, the speaker's position (placed, held, or transitional) can be determined based on fluctuation characteristics. Determining that the speaker is in a held state may include: increasing the confidence level of the held state when the fluctuation characteristics exceed a held state threshold for multiple consecutive sliding time windows; and outputting the held state when the confidence level reaches a held state confirmation threshold.
[0045] In step S104, auxiliary judgment information can be formed by combining touch activity, button activity, human proximity event, grip event, posture perturbation matching handshake mode, playback volume level, playback content type, or historical state information.
[0046] In decision node D105, the placement / handheld state determination module 104 determines whether the fluctuation characteristics or auxiliary determination information meet the handheld condition. When the fluctuation characteristics exceed the handheld threshold for multiple consecutive sliding time windows, or when touch activity, button activity, proximity to a human body event, gripping event, or posture perturbation consistent with the handheld shaking mode is detected, step S106 is entered, and the handheld state is output. The handheld state confidence can then be increased. When the handheld state confidence reaches a preset handheld confirmation threshold (e.g., 0.7 or 0.8), the speaker device is confirmed to be in a handheld state.
[0047] In decision node D107, the placement / handheld state determination module 104 determines whether the fluctuation characteristics continuously meet the placement stability conditions. When the fluctuation characteristics are below the placement threshold and continue for more than a preset placement confirmation time, and no touch activity, button activity, gripping event, or proximity to a human body event is detected, step S108 is entered, and the placement state is output. Then, the handheld state confidence is reduced (or the placement state confidence is increased). The placement confirmation time can be 1 second to 5 seconds (e.g., 2 seconds or 3 seconds) to ensure that the device has been stably placed on the table.
[0048] If neither the handheld confirmation condition nor the placement confirmation condition is met, proceed to step S109 and output a transition state. The current equalizer parameters can be maintained or parameter interpolation can be performed. Here, the specific meaning and calculation of "parameter interpolation" are explained: In transition state S109, the audio equipment has neither reached the low-fluctuation confirmation condition for stable placement nor the confidence threshold for complete handheld operation. At this time, the current equalizer parameters are maintained, or dynamic weighted interpolation processing is performed on the equalizer parameters for the placement state and the handheld state. Specifically, parameter interpolation refers to performing linear or smooth curve transition interpolation calculations between the baseline of the equalizer parameters for the placement state and the baseline of the equalizer parameters for the handheld state, based on the currently calculated real-time handheld state confidence (as the interpolation weight factor α). This allows the gain and filter coefficients of each frequency band of the equalizer to change dynamically and gradually with the confidence level, avoiding abrupt changes in sound due to the uncertain state. The concept and processing method of parameter interpolation are clear and explicit.
[0049] In some examples, the placement / handheld state determination module 104 may include a confidence accumulation unit and a hysteresis determination unit. The hysteresis determination unit uses different placement and handheld thresholds to form a hysteresis determination interval (the handheld threshold is higher than the placement threshold). The confidence accumulation unit updates the handheld state confidence and placement state confidence based on the fluctuation characteristics of a continuous sliding time window. During state switching, frequent switching of the output state caused by sensor noise is suppressed by continuous state confirmation or state holding time. The aforementioned thresholds, confidence levels, and time parameters can all be determined by hardware calibration.
[0050] like Figure 3 As shown, in step S110, the audio parameter control module 106 can select a target equalizer parameter between the placement state equalizer parameter and the handheld state equalizer parameter based on a determined state. The equalizer parameter library 105 stores at least the placement state equalizer parameter and the handheld state equalizer parameter. The handheld state equalizer parameter is used to compensate for acoustic response changes caused by near-field hearing, handholding obstruction, and human body coupling. The handheld state equalizer parameter includes one or more of the following: low-frequency compensation parameter, low-frequency limiting parameter, vocal clarity parameter, and near-field comfort parameter. The audio parameter control module 106 can select the corresponding handheld state equalizer parameter based on the holding posture, the possibility of speaker vent obstruction, the playback volume level, the type of playback content, or the microphone feedback result.
[0051] Here, the definitions of the various handheld equalizer parameters are explained: Low-frequency compensation parameters are used to appropriately enhance low-frequency extension when both hands are suspended in the air and the speakers are not obstructed; Low-frequency limiting parameters are used to attenuate low frequencies, such as 100Hz to 250Hz, when the device is close to the body or played at high volume, to prevent near-field resonance muddiness; Vocal clarity parameters are used to boost the gain in the 1kHz to 3kHz frequency band during voice or video calls; Near-field comfort parameters are used to attenuate the high-frequency harshness in the 4kHz to 8kHz frequency band and suppress transient sound pressure; Obstruction compensation parameters are used to increase high-frequency diffusion gain when a hand is detected obstructing the sound outlet. Each parameter type has clearly defined acoustic tuning standards for its frequency band and gain.
[0052] In step S111, during the audio signal processing of the audio equipment, the audio signal can be processed according to the target equalizer parameters. Specifically, the audio signal processing module 107 can perform multi-band digital filtering, dynamic range compression, and gain adjustment on the input audio signal according to the determined target equalizer parameters, and output the processed audio signal to the speaker output module 108 for electroacoustic conversion and sound generation. During audio playback, the audio signal processing module 107 can also provide feedback on the current playback volume level and audio spectrum characteristics to the placement / handheld status determination module 104 to assist in status determination.
[0053] In step S112, when the state switches from the placement state to the handheld state or vice versa, the smooth transition module smoothly transitions the equalizer parameters. Within a preset transition time, the smooth transition module interpolates the equalizer parameters for the placement state and the handheld state based on the confidence level of the handheld state (or the placement state) to obtain the current equalizer parameters. The current equalizer parameters are calculated according to the formula: Current equalizer parameter (EQ) = (1-α) × Placement state equalizer parameter + α × Handheld state equalizer parameter, where α is an interpolation coefficient between 0 and 1. The preset transition time can be from 50 milliseconds to 1000 milliseconds (e.g., 100 milliseconds, 300 milliseconds, or 500 milliseconds). A preset transition time that is too short (less than 50 milliseconds) is prone to producing an audible clicking sound, while a preset transition time that is too long (greater than 1000 milliseconds) will cause acoustic response lag. The parameter setting of 50 milliseconds to 1000 milliseconds ensures smooth and continuous timbre and timely response.
[0054] like Figure 4 As shown, the audio equipment is initially in a placed state, with low-amplitude gravity fluctuations. The confidence level for the handheld state is below the threshold, and the equalizer parameters for the placed state are output. When the user picks up the audio equipment, the gravity fluctuations become continuous and subtle, the confidence level for the handheld state gradually increases, and the equalizer parameters smoothly switch to the handheld state equalizer parameters. When the user holds the device and plays it near the field, the fluctuations remain within the handheld detection range, and the equalizer parameters for both the handheld and handheld states are maintained. After the user places the device back on a stable surface, the fluctuations decrease, the confidence level for the handheld state gradually decreases, and the equalizer parameters smoothly recover to the placed state equalizer parameters.
[0055] In one embodiment, the electronic device can be a portable Bluetooth speaker. The electronic device may include a processor, memory, a gravity sensor (triaxial gravity sensor), a speaker (speaker unit and amplifier module), a key input module, and a computer program stored in the memory. The memory pre-stores placement state equalizer parameters and handheld state equalizer parameters. The processor reads and executes the program to implement the method described above.
[0056] In another embodiment, the electronic device can be a mobile display device with a speaker (e.g., a digital photo frame, electronic calendar, portable screen, tablet computer, etc.). When the device is in a stand-mounted position, it uses equalizer parameters suitable for far-field viewing. When the user picks up the device and brings it close to their body to watch the video, the system determines the holding state through gravity fluctuations, touch activity, and posture changes, and switches to equalizer parameters suitable for near-field viewing to enhance dialogue clarity, reduce high-frequency stimulation at close range, and control instantaneous loudness.
[0057] In another embodiment, this disclosure provides a computer-readable storage medium having a computer program stored thereon. The computer-readable storage medium may be a non-volatile memory, a mobile memory, or a server storage medium. When the computer program is executed by a processor, it implements the methods described above.
[0058] The technical solutions disclosed herein are not limited to the embodiments described above. Any adjustments made to the types of fluctuation characteristics, auxiliary information, threshold setting methods, confidence update methods, equalizer parameter groups, smooth transition methods, and audio equipment types, as long as they do not depart from the technical concept of this disclosure, shall be considered to fall within the protection scope of this disclosure.
Claims
1. A method for handheld status recognition and adaptive equalizer switching of audio equipment, characterized in that, include: Collect signals from the gravity sensor of the audio equipment; Within a preset sliding time window, the gravity sensor signal is filtered and fluctuation characteristics are extracted; Based on the wave characteristics, it is determined whether the audio device is in a placed state, a handheld state, or a transitional state. Based on the determined state, select the target equalizer parameter between the placement state equalizer parameter and the handheld state equalizer parameter; During the audio signal processing of the audio equipment, the audio signal is processed according to the target equalizer parameters; and When switching from the placed state to the handheld state or vice versa, the equalizer parameters are smoothly transitioned.
2. The method according to claim 1, characterized in that, The gravity sensor signals include X-axis data, Y-axis data, and Z-axis data output from a triaxial gravity sensor, a triaxial accelerometer, or an inertial measurement unit.
3. The method according to claim 1, characterized in that, The fluctuation characteristics include one or more of the following: peak-to-peak value, variance, standard deviation, root mean square change, short-time energy, rate of change of direction, difference between adjacent sliding time windows, low-frequency jitter component, and frequency band energy.
4. The method according to claim 1, characterized in that, Determining that the audio device is in a handheld state includes: increasing the confidence level of the handheld state when the fluctuation feature is higher than the handheld threshold for multiple consecutive sliding time windows; and outputting the handheld state when the confidence level of the handheld state reaches the handheld confirmation threshold.
5. The method according to claim 1, characterized in that, Determining that the audio device is in a placement state includes: outputting the placement state when the fluctuation characteristics are below the placement threshold and continue for more than the placement confirmation time, and no touch activity, button activity, gripping event or human proximity event is detected.
6. The method according to claim 1, characterized in that, Determining whether the audio device is in a handheld state is also based on one or more auxiliary information from touch activity, button activity, proximity to human body events, gripping events, posture perturbation conforming to handheld shaking mode, playback volume level, playback content type, and historical status information.
7. The method according to claim 1, characterized in that, When determining whether the audio device is in a placed state, a handheld state, or a transitional state, different placement thresholds and handheld thresholds are used to form a hysteresis determination, and frequent state switching caused by sensor noise is suppressed by continuous state confirmation or state holding time.
8. The method according to claim 1, characterized in that, The handheld equalizer parameters include one or more of the following: low-frequency compensation parameters, low-frequency limiting parameters, vocal clarity parameters, and near-field comfort parameters. The corresponding handheld equalizer parameters are selected based on the holding posture, the possibility of speaker vent obstruction, the playback volume level, the type of playback content, or the microphone feedback results.
9. The method according to claim 1, characterized in that, The smooth transition includes: within a preset transition time, interpolating the equalizer parameters of the placement state and the equalizer parameters of the handheld state based on the confidence level of the handheld state or the confidence level of the placement state to obtain the current equalizer parameters; the preset transition time is from 50 milliseconds to 1000 milliseconds.
10. A device for recognizing the handheld status of an audio device and adaptively switching the equalizer, characterized in that, include: Gravity sensor acquisition module, used to acquire gravity sensor signals from audio equipment; The wave feature extraction module is used to filter the gravity sensor signal and extract wave features within a preset sliding time window; The state determination module is used to determine whether the audio device is in a placed state, a handheld state, or a transitional state based on the fluctuation characteristics. The audio parameter control module is used to select the target equalizer parameter between the placement state equalizer parameter and the handheld state equalizer parameter based on the determined state. An audio signal processing module is used to process the audio signal according to the target equalizer parameters during the audio signal processing process of the audio equipment. as well as The smooth transition module is used to smoothly transition the equalizer parameters when switching from the placed state to the handheld state or vice versa.
11. The apparatus according to claim 10, characterized in that, It also includes an auxiliary information acquisition module, which is used to collect one or more auxiliary information from touch activity, button activity, human proximity event, grip event, playback volume level, playback content type or historical status information, and the status determination module also determines the audio device to be in a placed state, a handheld state or a transitional state based on the auxiliary information.
12. The apparatus according to claim 10, characterized in that, The state determination module includes a confidence accumulation unit and a hysteresis determination unit. The confidence accumulation unit is used to update the confidence of the handheld state and the confidence of the placement state according to the fluctuation characteristics of the continuous sliding time window. The hysteresis determination unit is used to output the state according to different handheld thresholds and placement thresholds.
13. An electronic device, characterized in that, The device includes a processor, a memory, a gravity sensor, a speaker, and a program stored in the memory, which, when executed by the processor, causes the electronic device to perform the method of any one of claims 1 to 9.
14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 9.
