Audio signal processing method, system and electronic device

Abstract

The application relates to the technical field of electronic circuits, and discloses an audio signal processing method, system and electronic device. The audio signal processing method of the application embodiment inputs an original audio signal and a first audio signal processed by a loudspeaker protection processing module into a perceptual metric model, performs perceptual difference analysis and processing on the original audio signal and the first audio signal by using the perceptual metric model, and obtains a perceptual metric score. Then, the perceptual metric score is output to a parameter control module, the parameter control module generates a control instruction based on the perceptual metric score and outputs the control instruction to the loudspeaker protection processing module, and the loudspeaker protection processing module outputs a target audio signal after adjusting and updating the original audio signal according to the control instruction. In the above manner, the influence of loudspeaker protection actions on audio quality or subjective listening can be minimized while protecting the loudspeaker from heat and mechanical damage.

Description

Technical Field

[0001] This application relates to the field of electronic circuit technology, and in particular to an audio signal processing method, system and electronic device. Background Technology

[0002] Loudspeakers (speakers) are devices that convert electrical signals into acoustic signals, and their performance and lifespan are often threatened by overload. Overload can be caused by excessively large input signals, leading to overheating of the voice coil or diaphragm displacement exceeding the linear range, resulting in irreversible damage or severe distortion. Therefore, loudspeaker protection technology has always been an important component of audio systems. Traditional loudspeaker protection methods are mainly based on the monitoring and control of physical parameters. For example, by monitoring the current or temperature of the voice coil and estimating the diaphragm displacement, the system protects the loudspeaker by reducing gain or limiting signal amplitude when these parameters exceed safe thresholds. While these methods can prevent hardware damage, they often come at the cost of sound quality, because protective actions (such as hard limiting or compression) introduce audible distortion or dynamic changes, affecting the listening experience. Summary of the Invention

[0003] To address the aforementioned technical problems, embodiments of this application provide an audio signal processing method, system, and electronic device.

[0004] In a first aspect, embodiments of this application provide an audio signal processing method, comprising: a speaker protection processing module and a perception measurement model respectively receiving an original audio signal; the speaker protection processing module adjusting the original audio signal in real time to obtain a first audio signal, and outputting the first audio signal to the perception measurement model; the perception measurement model performing perception difference analysis processing on the original audio signal and the first audio signal to obtain a perception measurement score, and outputting the perception measurement score to a parameter control module; the parameter control module generating a control command based on the perception measurement score and outputting it to the speaker protection processing module; and the speaker protection processing module adjusting and updating the original audio signal according to the control command and then outputting a target audio signal.

[0005] It is understood that the audio signal processing method of this application embodiment inputs the original audio signal and the first audio signal processed by the speaker protection processing module into a perceptual measurement model. The perceptual measurement model is then used to perform perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score. This perceptual measurement score is then output to a parameter control module. The parameter control module generates control commands based on the perceptual measurement score and outputs them to the speaker protection processing module. The speaker protection processing module adjusts and updates the original audio signal according to the control commands and outputs the target audio signal. Through this method, the loss of audio quality caused by speaker protection actions can be reduced, thereby improving the subjective listening experience. In this way, while protecting the speaker from heat and mechanical damage, the impact of speaker protection actions on audio quality or subjective listening experience can be minimized.

[0006] In some possible implementations of the first aspect mentioned above, the perceptual measurement model includes a pre-trained psychoacoustic model; the perceptual measurement model performs perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score, including: using the pre-trained psychoacoustic model to determine the first auditory feature information of the original audio signal, the second auditory feature information of the first audio signal, and the distortion information between the original audio signal and the first audio signal, and determining the perceptual measurement score based on the first auditory feature information, the second auditory feature information, and the distortion information, the perceptual measurement score including auditory quality scores under different time dimensions.

[0007] In some possible implementations of the first aspect mentioned above, the perceptual measurement score includes a first time dimension score, a second time dimension score, and a third time dimension score; wherein, the first time dimension score reflects the instantaneous listening quality, the second time dimension score reflects the listening quality within a preset duration, and the third time dimension score reflects the long-term listening quality.

[0008] In some possible implementations of the first aspect above, the speaker protection processing module obtains a first audio signal by adjusting the original audio signal in real time, including: the speaker protection processing module inputs the original audio signal into a displacement protection model; the displacement protection model estimates the diaphragm displacement of the speaker based on the original audio signal and outputs a first displacement protection parameter; the speaker protection processing module obtains the first audio signal by adjusting the original audio signal in real time based on the first displacement protection parameter.

[0009] In some possible implementations of the first aspect above, the control command includes a second displacement protection parameter; the parameter control module generates a control command based on the perception metric score and outputs it to the loudspeaker protection processing module, including: the parameter control module adjusts the first displacement protection parameter based on the first time dimension score and / or the second time dimension score to obtain the second displacement protection parameter; the parameter control module adjusts the model parameters of the displacement protection model based on the third time dimension score; the parameter control module outputs the second displacement protection parameter and the model parameters of the displacement protection model to the loudspeaker protection processing module.

[0010] In some possible implementations of the first aspect above, the speaker protection processing module obtains a first audio signal by adjusting the original audio signal in real time, including: the speaker protection processing module inputs the original audio signal into a temperature protection model; the temperature protection model estimates the voice coil temperature of the speaker based on the original audio signal and outputs a first temperature protection parameter; the speaker protection processing module obtains the first audio signal by adjusting the original audio signal in real time based on the first temperature protection parameter.

[0011] In some possible implementations of the first aspect above, the control command includes a second temperature protection parameter; the parameter control module generates a control command based on the perception metric score and outputs it to the speaker protection processing module, including: the parameter control module adjusts the first temperature protection parameter based on the second time dimension score to obtain the second temperature protection parameter; the parameter control module adjusts the model parameters of the temperature protection model based on the third time dimension score; the parameter control module outputs the second temperature protection parameter and the model parameters of the temperature protection model to the speaker protection processing module.

[0012] In some possible implementations of the first aspect mentioned above, the control command further includes a target pre-compensation gain parameter; the parameter control module generates control commands based on the perception metric score and outputs them to the speaker protection processing module, and further includes: the parameter control module adjusts the original pre-compensation gain parameter based on the third time dimension score to obtain the target pre-compensation gain parameter; wherein, the target pre-compensation gain parameter is used to adjust the global gain or the gain of a specific frequency point of the original audio signal.

[0013] Secondly, embodiments of this application provide an audio signal processing system, including: a speaker protection processing module, configured to receive an original audio signal, adjust the original audio signal in real time to obtain a first audio signal, and output the first audio signal to a perception measurement model; a perception measurement model, configured to receive the original audio signal and the first audio signal, perform perception difference analysis processing on the original audio signal and the first audio signal to obtain a perception measurement score, and output the perception measurement score to a parameter control module; a parameter control module, configured to generate control commands based on the perception measurement score and output them to the speaker protection processing module; and a speaker protection processing module, further configured to adjust and update the original audio signal according to the control commands and output a target audio signal.

[0014] Thirdly, embodiments of this application provide an electronic device including the audio signal processing system and the speaker described in the second aspect above, wherein the audio signal processing system and the speaker are electrically connected and are used to send a target audio signal to the speaker.

[0015] It is understandable that the technical effects of the second and third aspects can be referenced from the technical effects of the first aspect, and will not be elaborated further. Attached Figure Description

[0016] Figure 1 According to some embodiments of this application, an architecture block diagram of an audio signal processing system is shown;

[0017] Figure 2 According to some embodiments of this application, a flowchart of an audio signal processing method is shown;

[0018] Figure 3 According to some embodiments of this application, a schematic diagram of the adjustment object corresponding to the sound quality score under different time dimensions is shown;

[0019] Figure 4 According to some embodiments of this application, a logic block diagram for audio signal processing is shown;

[0020] Figure 5 According to some embodiments of this application, a principle block diagram of adjusting protection parameters using a sensing measurement model is shown;

[0021] Figure 6 According to some embodiments of this application, a schematic diagram of the structure of an electronic device 100 is shown. Detailed Implementation

[0022] The illustrative embodiments of this application include, but are not limited to, an audio signal processing method, system, and electronic device. The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0023] As mentioned earlier, traditional speaker protection methods (such as hard limiting or compression) negatively impact the listening experience. With the development of digital signal processing technology, more sophisticated protection strategies have emerged. For example, multi-band dynamic processing is used to compress or limit only in at-risk frequency bands, or look-ahead buffering is used to smoothly apply protection and reduce audible noise. However, these methods still rely on the engineer's experience to set processing parameters and lack a quantitative assessment of the auditory impact of the processing results.

[0024] Based on this, this application provides an audio signal processing method. The method involves inputting the original audio signal and a first audio signal processed by a speaker protection module into a perceptual measurement model. The perceptual measurement model is then used to perform perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score. This perceptual measurement score is then output to a parameter control module. The parameter control module generates control commands based on the perceptual measurement score and outputs them to the speaker protection module. The speaker protection module adjusts and updates the original audio signal according to the control commands and outputs the target audio signal. This approach reduces the loss of audio quality caused by speaker protection actions, thereby improving subjective listening experience. In this way, the impact of speaker protection actions on audio quality or subjective listening experience can be minimized while protecting the speaker from heat and mechanical damage.

[0025] The audio signal processing method of this application embodiment will be further described below with reference to the accompanying drawings.

[0026] Before introducing the audio signal processing method of the embodiments of this application, the application scenarios of the audio processing method of the embodiments of this application will be explained first. Figure 1 An architectural block diagram of an audio signal processing system is shown.

[0027] like Figure 1 As shown, the audio signal processing system of this application embodiment includes an audio input module 101, a speaker protection processing module 102, a perception measurement model 103, a parameter control module 104, and a speaker 105. The output terminal of the audio input module 101 is connected to the input terminals of the speaker protection processing module 102 and the perception measurement model 103. The output terminal of the speaker protection processing module 102 is connected to the speaker 105 and the input terminal of the perception measurement model 103. The output terminal of the perception measurement model 103 is connected to the input terminal of the parameter control module 104, and the output terminal of the parameter control module 104 is connected to the input terminal of the speaker protection processing module 102.

[0028] The audio input module 101 is used to output the original audio signal; the speaker protection processing module 102 is used to protect the audio signal, for example, by protecting the audio signal according to the speaker protection parameters; the perception measurement model 103 includes a perception measurement model for perceptual analysis of the audio signal; the parameter control module 104 is used to adjust the protection parameters of the speaker protection processing module 102; and the speaker 105 is used to output the audio signal.

[0029] Figure 2 A flowchart illustrating an audio signal processing method in some embodiments of this application is shown. (In conjunction with...) Figure 1 and Figure 2 As shown, the audio signal processing method includes the following steps:

[0030] S201: The speaker protection processing module 102 and the perception measurement model respectively receive the original audio signal.

[0031] It can be understood that the audio input module 101 outputs the original audio signal to the speaker protection processing module 102 and the perception measurement model 103 respectively, so that the speaker protection processing module 102 and the perception measurement model 103 receive the original audio signal respectively.

[0032] S202: The speaker protection processing module 102 adjusts the original audio signal in real time to obtain the first audio signal, and outputs the first audio signal to the perception measurement model.

[0033] It can be understood that the speaker protection processing module 102 adjusts the original audio signal in real time to obtain the first audio signal. Specifically, the first audio signal can be the audio signal after the speaker protection processing module 102 performs protection processing on the original audio signal based on speaker protection parameters.

[0034] Protection measures refer to actions taken to protect the loudspeaker from heat and mechanical damage, including but not limited to temperature protection measures and / or displacement protection measures. Temperature protection measures may involve processing the original audio signal to prevent the loudspeaker's voice coil from overheating. Displacement protection measures may involve processing the original audio signal to prevent the physical movement of the loudspeaker's diaphragm from exceeding the safe range of its mechanical structure. Accordingly, loudspeaker protection parameters may include, but are not limited to, temperature protection parameters and / or displacement protection parameters.

[0035] S203: The perceptual measurement model performs perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score, and outputs the perceptual measurement score to the parameter control module.

[0036] It is understandable that the perceptual measurement model is used to receive the first audio signal and the original audio signal, and to quantitatively evaluate the subjective listening quality of the audio signal from the perspective of sound quality. For example, it assesses whether the dynamic range is over-compressed, whether the low-frequency impact is lost, whether the distortion is audible, etc., and performs perceptual difference analysis to output a perceptual measurement score.

[0037] Dynamic range refers to the ratio between the maximum and minimum sound pressure level over a period of time. Low-frequency impact refers to the audible characteristics of low-frequency transient signals. Whether distortion is audible refers to whether the nonlinear changes introduced during audio processing can be perceived by the human auditory system.

[0038] In some embodiments, the perceptual metric score may include sound quality scores across different time dimensions. For example, the perceptual metric score may include short-term, medium-term, and long-term evaluation scores. The short-term evaluation score reflects instantaneous changes in sound quality, the medium-term evaluation score reflects sound quality trends over a period of time, and the long-term evaluation score reflects long-term sound quality balance and system protection effectiveness. In other words, the perceptual metric model can achieve sound quality risk assessment across multiple time scales (short-term, medium-term, and long-term).

[0039] S204: The parameter control module 104 generates control commands based on the perception measurement score and outputs them to the speaker protection processing module 102.

[0040] It is understood that this application evaluates the impact of protection actions on sound quality through a perceptual measurement model, and generates control commands to adjust speaker protection parameters based on the perceptual measurement score output by the perceptual measurement model.

[0041] As mentioned above, speaker protection parameters may include, but are not limited to, temperature protection parameters and / or displacement protection parameters. Correspondingly, control commands may include, but are not limited to, temperature adjustment parameters and / or displacement adjustment parameters. The parameter control module 104 generates control commands based on the perception metric score, including generating adjustment commands for temperature protection parameters and / or displacement protection parameters based on the perception metric score.

[0042] It is understandable that the corresponding perception metric score can include hearing quality scores across different time dimensions, such as short-term, mid-term, and long-term evaluation scores. Accordingly, temperature protection parameters or displacement protection parameters can be adjusted based on the scores corresponding to different time dimensions. For example, temperature protection parameters can be adjusted based on the mid-term evaluation score. The specific adjustment methods will be explained in detail below, and will not be elaborated upon here.

[0043] S205: The speaker protection processing module 102 adjusts and updates the original audio signal according to the control command and then outputs the target audio signal.

[0044] It is understandable that the temperature adjustment parameters and / or displacement adjustment parameters carried by the control command are protection parameters adjusted based on the perception metric score. Therefore, after processing the original audio signal based on the adjusted protection parameters, the target audio signal obtained can ensure the physical safety of the speaker while minimizing the impact of the protection action on the subjective listening experience.

[0045] In summary, the audio signal processing method provided in this application inputs the original audio signal and the pre-processed audio signal (first audio signal) into a perceptual measurement model. The perceptual measurement model then performs perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score. Based on this score, the speaker protection parameters are adjusted. The adjusted parameters reduce the loss of audio quality caused by speaker protection actions, thereby improving subjective listening experience. Thus, the audio signal processing method in this application can minimize the impact of speaker protection actions on audio quality or subjective listening experience while protecting the speaker from heat and mechanical damage.

[0046] In some embodiments, the perceptual measurement model described above includes a pre-trained psychoacoustic model. It is understood that the pre-trained psychoacoustic model can be trained based on algorithms such as Perceptual Evaluation of Speech Quality (PESQ) or Perceptual Objective Listening Quality Analysis (POLQA), or other neural network algorithms / models.

[0047] Accordingly, step S203 above, in which the perceptual measurement model performs perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score, may include: using a pre-trained psychoacoustic model to determine the first auditory feature information of the original audio signal, the second auditory feature information of the first audio signal, and the distortion information between the original audio signal and the first audio signal, and determining the perceptual measurement score based on the first auditory feature information, the second auditory feature information, and the distortion information, wherein the perceptual measurement score includes auditory quality scores under different time dimensions.

[0048] It is understandable that the pre-trained psychoacoustic model, after performing feature extraction, difference quantization, and multi-timescale analysis on the original audio signal and the first audio signal, outputs a listening quality score at different time dimensions. The first / second listening feature information may include, but is not limited to, loudness, dynamic range, and spectral characteristics. Distortion information includes, but is not limited to, harmonic distortion, intermodulation distortion, and dynamic range loss.

[0049] Perceptual metric scores include auditory quality scores across different time dimensions, facilitating more refined control. In some embodiments, perceptual metric scores include first time dimension scores, second time dimension scores, and third time dimension scores.

[0050] The first-time dimension score corresponds to the short-term assessment score mentioned above, reflecting instantaneous auditory quality, such as the impact of a drumbeat or the clarity of a note; it can also reflect instantaneous risk (such as whether a loud drum sound will cause the voice coil to bottom out). The assessment window is in milliseconds, for example, 10ms to 100ms.

[0051] The second time dimension score corresponds to the mid-term evaluation score mentioned above, reflecting the listening quality within a preset time period, such as the energy and brightness of a piece of music; it can also reflect trend risks (such as a steady increase in temperature due to continuous high volume). The evaluation window is in the second range, for example, a few seconds to a dozen seconds, or 1 second to 30 seconds, or a musical passage or a continuous conversation.

[0052] The third time dimension score corresponds to the long-term assessment score mentioned above, reflecting long-term listening quality, such as whether the overall listening experience is balanced; it can also reflect cumulative risk and whether the protection strategy has maintained acceptable sound quality over a long period of time. The assessment window can range from minutes to hours, or even span multiple usage sessions.

[0053] It is understandable that traditional speaker protection actions (such as limiting and compression) are abrupt and audible, resulting in disrupted music dynamics, abrupt changes in timbre, and a poor user experience. However, in this embodiment, the impact of protection actions on sound quality is evaluated in real time using the aforementioned perception measurement model. Based on this (sound quality score at different time dimensions), speaker protection parameters are adjusted, which reduces the impact of protection actions on sound quality. This makes the sound quality degradation almost imperceptible to the user when protection is triggered, greatly improving the listening experience under high load.

[0054] In some embodiments, step 202 above, in which the speaker protection processing module 102 adjusts the original audio signal in real time to obtain the first audio signal, includes: the speaker protection processing module 102 inputs the original audio signal into a displacement protection model; the displacement protection model estimates the diaphragm displacement of the speaker based on the original audio signal and outputs a first displacement protection parameter; the speaker protection processing module 102 adjusts the original audio signal in real time based on the first displacement protection parameter to obtain the first audio signal.

[0055] It can be understood that a displacement protection model can be a mathematical or algorithmic model used to estimate the displacement of a loudspeaker diaphragm. Specifically, the displacement protection model is primarily used to estimate the displacement of the loudspeaker diaphragm based on the original audio signal (or its characteristics, such as low-frequency energy). Specifically, the displacement protection model is used to estimate the original audio signal, and combined with a preset safe displacement threshold, to determine the first displacement protection parameter used to prevent mechanical overload (bottoming / edge impact). The first displacement protection parameter includes, but is not limited to: fast-response limiter parameters, dynamic limiter (compressor) parameters for the full frequency band or low frequency band, dynamic high-pass filter parameters, and crossover control parameters.

[0056] In some embodiments, step 202 above, in which the speaker protection processing module 102 adjusts the original audio signal in real time to obtain the first audio signal, includes: the speaker protection processing module 102 inputs the original audio signal into a temperature protection model; the temperature protection model estimates the voice coil temperature of the speaker based on the original audio signal and outputs a first temperature protection parameter; and the speaker protection processing module 102 adjusts the original audio signal in real time based on the first temperature protection parameter to obtain the first audio signal.

[0057] It can be understood that a temperature protection model is a mathematical model used to estimate the speaker voice coil temperature. Specifically, the temperature protection model is primarily used to estimate the speaker voice coil temperature based on information such as the power of the original audio signal. Specifically, the temperature protection model is used to estimate the temperature of the original audio signal, and combined with a preset safe temperature threshold, to determine the first temperature protection parameter used to prevent thermal damage. The first temperature protection parameter includes, but is not limited to: compressor threshold, compression ratio, start / release time, frequency equalization curve, etc.

[0058] In some embodiments, the speaker protection parameters also include pre-compensation gain parameters. Pre-compensation gain parameters include, but are not limited to, the main output gain value and the equalizer frequency gain setting. The main output gain value is used for global volume fine-tuning, while the equalizer frequency gain setting is used to achieve gain at specific frequencies to optimize the final output sound or meet specific acoustic goals.

[0059] The audio signal processing method of this application embodiment can dynamically adjust temperature protection parameters, displacement protection parameters, and pre-compensation gain parameters based on the listening quality scores (first time dimension score, second time dimension score, and third time dimension score) under different time dimensions.

[0060] Combination Figure 3As shown, the first-time dimension score (short-term evaluation score) is used to determine the degree of damage to instantaneous sound quality. Therefore, the first displacement protection parameter can be adjusted based on the first-time dimension score. Here, the first displacement protection parameter can be adjusted directly, or it can be adjusted by adjusting the model parameters of the displacement protection model.

[0061] The second time dimension score (mid-term evaluation score) reflects the cumulative degree of ongoing risk, such as a continuous rise in temperature or a prolonged high displacement leading to an overall degradation trend in sound quality. Therefore, the first displacement protection parameter and / or the first temperature protection parameter can be adjusted based on the second time dimension score. Similarly, the first temperature protection parameter can be adjusted directly, or by adjusting the model parameters of the temperature protection model.

[0062] The third-time dimension score (long-term evaluation score) reflects whether the system's overall protection strategy is too conservative or aggressive, and whether it aligns with user preferences or specific environments. Therefore, the default parameters (model parameters, such as safe temperature threshold and safe displacement threshold) of the temperature protection model and displacement protection model, as well as the first pre-compensation gain parameter, can be adjusted based on the third-time dimension score to optimize the long-term system protection strategy. This enables both adaptive and learning mechanisms.

[0063] In some embodiments, the control command includes a second displacement protection parameter, which is an adjusted version of the first displacement protection parameter. The specific content can be referenced from the first displacement protection parameter. Specifically, in S204 above: the parameter control module 104 generates a control command based on the perception metric score and outputs it to the speaker protection processing module 102, including: the parameter control module 104 adjusting the first displacement protection parameter based on the first time dimension score and / or the second time dimension score to obtain the second displacement protection parameter; the parameter control module 104 adjusting the model parameters of the displacement protection model based on the third time dimension score; and the parameter control module 104 outputting the second displacement protection parameter and the model parameters of the displacement protection model to the speaker protection processing module 102.

[0064] Understandably, this is usually caused by a large, transient signal. Under the influence of displacement control, this leads to momentary disruption of dynamics or spectral fragmentation. This momentary audio quality degradation is reflected in the first time dimension score, which can then be used to adjust the first displacement protection parameters. For example, the displacement protection parameters can be fine-tuned to reduce the impact of these two factors on audio dynamics. Furthermore, the overall audio quality degradation trend caused by a persistently high displacement level will be reflected in the second time dimension score, which can also be used to adjust the first displacement protection parameters.

[0065] In some embodiments, the control command includes a second temperature protection parameter, which is an adjusted version of the first temperature protection parameter. The specific content can be referenced from the first temperature protection parameter. Specifically, in S204 above: the parameter control module 104 generates a control command based on the perception measurement score and outputs it to the speaker protection processing module 102, including: the parameter control module 104 adjusting the first temperature protection parameter based on the second time dimension score to obtain the second temperature protection parameter; the parameter control module 104 adjusting the model parameters of the temperature protection model based on the third time dimension score; and the parameter control module 104 outputting the second temperature protection parameter and the model parameters of the temperature protection model to the speaker protection processing module 102.

[0066] It's understandable that a sustained increase in temperature can lead to a general decline in sound quality, such as continuous dynamic compression (flattening of the sound), decreased brightness (thermal compression), or a slow increase in distortion. This can be reflected in the second time dimension score. Therefore, the first temperature protection parameters can be adjusted based on the second time dimension score. For example, proactive and smooth trend management can be implemented to prevent the system from entering a dangerous or severely degraded state, achieving a smooth transition.

[0067] In some embodiments, the control command further includes a target pre-compensation gain parameter; specifically, the above-mentioned S204: the parameter control module 104 generates a control command based on the perception metric score and outputs it to the speaker protection processing module 102, including: the parameter control module adjusts the original pre-compensation gain parameter based on the third time dimension score to obtain the target pre-compensation gain parameter; wherein, the target pre-compensation gain parameter is used to adjust the global gain or the gain of a specific frequency point of the original audio signal.

[0068] As mentioned earlier, the third-time dimension score reflects whether the system's overall protection strategy is too conservative or aggressive, and whether it aligns with user preferences or specific environments. Therefore, the default parameters (model parameters, such as safe temperature threshold and safe displacement threshold) of the temperature protection model and displacement protection model, as well as the first pre-compensation gain parameter, can be adjusted based on the third-time dimension score. The third-time dimension score, which does not address instantaneous issues but optimizes fundamental system settings, is a key element in achieving intelligent self-adaptation in speaker protection systems.

[0069] In related technologies, the model parameters (baseline) of temperature protection models, the model parameters of displacement protection models, or the pre-compensation gain parameters are fixed at the factory, which cannot adapt to the diverse user habits, environmental conditions, and device aging. In the embodiments of this application, the third time dimension scoring can drive the optimization of baseline parameters, enabling the speaker protection system to continuously learn. If the current strategy is found to be too conservative (protection is rarely triggered and the sound quality is good), the threshold release performance is automatically relaxed (e.g., loudness is increased); if it is too aggressive (protection is frequently triggered), the parameters are tightened to ensure safety.

[0070] In some embodiments, a persistently high second-time dimension score indicates a significant medium-term risk. Therefore, the reasons for the persistently high second-time dimension score can be analyzed by combining the temperature estimated by the temperature protection model and the displacement estimated by the displacement protection model. For example, if the temperature estimated by the temperature protection model remains high while the displacement estimated by the displacement protection model is relatively normal, it can be determined that the persistently high second-time dimension score is due to temperature rise, allowing for more precise adjustment of the temperature protection parameters. Alternatively, the displacement protection parameters can be adjusted simultaneously with the temperature protection parameters. This way, when the second-time dimension score indicates an excessively rapid temperature rise, the displacement protection parameters can be proactively tightened simultaneously to prevent the suspension from softening under high temperatures.

[0071] Compared to related technologies, temperature protection and displacement protection typically operate independently with fixed parameters, which can easily lead to conflicts (e.g., reducing gain to lower the temperature may result in insufficient low-frequency response, forcing a boost and potentially causing displacement risks). In this embodiment, a joint decision based on scores from different time dimensions allows for dynamic coordination of temperature protection and displacement protection strategies. This enables system-level optimization, allowing the system to withstand both sustained high temperatures and sudden shocks in more complex usage scenarios, thus enhancing system robustness.

[0072] Combination Figure 4 As shown, Figure 4 A logic block diagram of audio signal processing according to an embodiment of this application is shown.

[0073] The input signal is the original audio signal. The speaker protection processing module 102 processes the original audio signal based on the original speaker protection parameters (such as the first displacement protection parameter and the first temperature protection parameter mentioned above) to obtain the first audio signal. The first audio signal is sent to the perception measurement model as the "signal to be evaluated". At the same time, the original audio signal input is also input to the perception measurement model as the "reference signal". The perception measurement model objectively compares the perceptual differences between the "reference signal" (original sound quality) and the "signal to be evaluated" (processed sound quality) and analyzes the impact of this difference on the listening experience (such as dynamic loss and increased distortion), and then outputs a quantified perception measurement score. The perception measurement model sends this score to the parameter control module for protection parameter control. For example, based on the perception measurement score, the parameter control module makes intelligent decisions and generates new control commands to dynamically adjust the protection parameters in the speaker protection processing module, such as the temperature protection parameter, displacement protection parameter, pre-compensation gain parameter, or model parameter mentioned above, thereby forming a real-time optimized closed loop. The working logic of the audio signal processing method is mainly divided into two parts: perception and diagnosis, and coordination and execution, which can realize intelligent closed-loop control with perception as the core.

[0074] In some embodiments, the speaker protection processing module 102 processes the raw audio signal based on protection parameters involved in the speaker protection system (such as temperature protection parameters, displacement protection parameters, pre-compensation gain parameters, or model parameters). The multi-timescale evaluation output by the perceptual metric model can be used to adjust the protection parameters involved in the speaker protection system.

[0075] Figure 5 A block diagram illustrating the principle of adjusting protection parameters using a sensing metric model is shown. Combined with... Figure 5 As shown, the loudspeaker protection system includes a pre-compensation gain subsystem, a temperature protection subsystem, and a displacement protection subsystem. The pre-compensation gain subsystem involves controlling the equalizer and the overall gain; the temperature protection subsystem uses a temperature protection model to determine temperature protection parameters; and the displacement protection subsystem uses a displacement protection model to determine displacement protection parameters. The multi-timescale evaluation output by the perception measurement model includes a first time-dimensional score (short-term evaluation score), a second time-dimensional score (medium-term evaluation score), and a third time-dimensional score (long-term evaluation score). The first time-dimensional score (short-term evaluation score) can be fed back to the displacement protection subsystem to adjust the displacement protection parameters or the model parameters of the displacement protection model; the second time-dimensional score (medium-term evaluation score) can be fed back to the temperature protection subsystem to adjust the temperature protection parameters or the model parameters of the temperature protection model; and the third time-dimensional score (long-term evaluation score) can be fed back to the pre-compensation gain subsystem to adjust the equalizer and the overall gain. It is understood that the specific adjustment methods can be found in the relevant description of step S204 in the above method embodiment, and will not be repeated here.

[0076] As can be seen, the embodiments of this application construct a perception measurement model based on three time scales: short-term (transient), medium-term (trend), and long-term (baseline). The evaluation results are then accurately fed back to the corresponding protection subsystems (displacement protection subsystem, temperature protection subsystem, and pre-compensation gain subsystem), forming a hierarchical and refined intelligent control closed loop. Furthermore, this method can automatically determine whether the current protection action is "conservative," "aggressive," or "appropriate" by analyzing the protection trigger frequency and sound quality score distribution during long-term operation. It can also safely and progressively recalibrate the model parameters of the temperature protection model and displacement protection model, and adjust the safety threshold baseline, achieving personalized continuous optimization. In addition, minimizing perceived sound quality loss is taken as a core optimization objective that is equally important as physical safety, realizing a paradigm shift from "protecting hardware" to "protecting user experience."

[0077] Furthermore, traditional speaker protection methods rely on expensive temperature sensors, displacement sensors, or highly consistent speaker unit parameters. In contrast, the audio signal processing method and system of this application, based on software algorithm models (sensory measurement model, temperature protection model, displacement protection model) and audio signal analysis, can compensate for hardware parameter deviations or sensor deficiencies through learning and adaptation.

[0078] This application also provides an audio signal processing system, such as... Figure 1 As shown, it includes at least a speaker protection processing module 102, a perception measurement model 103, and a parameter control module 104.

[0079] The speaker protection processing module 102 is used to receive the original audio signal, adjust the original audio signal in real time to obtain the first audio signal, and output the first audio signal to the perception measurement model; the perception measurement model 103 is used to receive the original audio signal and the first audio signal, perform perception difference analysis processing on the original audio signal and the first audio signal to obtain a perception measurement score, and output the perception measurement score to the parameter control module; the parameter control module 104 is used to generate control commands based on the perception measurement score and output them to the speaker protection processing module; the speaker protection processing module 102 is also used to adjust and update the original audio signal according to the control commands and output the target audio signal.

[0080] Understandably, in practical applications, audio signal processing systems can be integrated in the form of chips.

[0081] This application also provides an electronic device, including an audio signal processing system and a speaker, wherein the audio signal processing system and the speaker are electrically connected and are used to send a target audio signal to the speaker.

[0082] Figure 6 According to some embodiments of this application, a schematic diagram of the structure of an electronic device 100 is shown. For example... Figure 6 As shown, the electronic device 100 includes one or more processors 1001, system memory 1002, non-volatile memory (NVM) 1003, communication interface 1004, input / output (I / O) device 1005, and system control logic 1006 for coupling the processor 1001, system memory 1002, non-volatile memory 1003, communication interface 1004, and input / output (I / O) device 1005. Wherein:

[0083] The processor 1001 may include one or more processing units, such as a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), micro-programmed control unit (MCU), artificial intelligence (AI) processor, or field programmable gate array (FPGA), neural network processing unit (NPU), etc. The data processing unit or processing circuit may include one or more single-core or multi-core processors.

[0084] System memory 1002 is volatile memory, such as random-access memory (RAM), double data rate synchronous dynamic random access memory (DDR SDRAM), etc. System memory is used for temporary storage of data and / or instructions.

[0085] The non-volatile memory 1003 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions. In some embodiments, the non-volatile memory 1003 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as a hard disk drive (HDD), a compact disc (CD), a digital versatile disc (DVD), a solid-state drive (SSD), etc. In some embodiments, the non-volatile memory 1003 may also be a removable storage medium, such as a secure digital (SD) memory card, etc.

[0086] Specifically, system memory 1002 and non-volatile memory 1003 may each include a temporary copy and a permanent copy of instruction 1007. Instruction 1007 may include, when executed by at least one of processors 1001, causing electronic device 100 to implement the audio signal processing methods provided in the embodiments of this application.

[0087] The communication interface 1004 may include a transceiver for providing a wired or wireless communication interface for the electronic device 100, thereby enabling communication with any other suitable device via one or more networks. In some embodiments, the communication interface 1004 may be integrated into other components of the electronic device 100, for example, the communication interface 1004 may be integrated into the processor 1001.

[0088] Input / output (I / O) device 1005 may include input devices such as a keyboard and mouse, and output devices such as a speaker and a monitor. Users can interact with electronic device 100 through input / output (I / O) device 1005. In this embodiment, the user can hear a first audio signal or a target audio signal output by electronic device 100 through a speaker.

[0089] System control logic 1006 may include any suitable interface controller to provide any suitable interface to other modules of electronic device 100. For example, in some embodiments, system control logic 1006 may include one or more memory controllers to provide an interface to system memory 1002 and non-volatile memory 1003.

[0090] In some embodiments, at least one of the processors 1001 may be packaged together with the logic of one or more controllers for system control logic 1006 to form a system in package (SiP). In other embodiments, at least one of the processors 1001 may also be integrated on the same chip with the logic of one or more controllers for system control logic 1006 to form a system-on-chip (SoC).

[0091] Understandable. Figure 6 The structure of the electronic device 100 shown is merely an example. In other embodiments, the electronic device 100 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0092] It should be noted that in the examples and description of this patent, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0093] Although this application has been illustrated and described with reference to certain preferred embodiments thereof, those skilled in the art will understand that various changes in form and detail may be made thereto without departing from the scope of this application.

Claims

1. An audio signal processing method, characterized in that, include: The speaker protection processing module and the perception measurement model respectively receive the original audio signal; The speaker protection processing module adjusts the original audio signal in real time to obtain a first audio signal, and outputs the first audio signal to the perception measurement model; The perception measurement model performs perception difference analysis on the original audio signal and the first audio signal to obtain a perception measurement score, and outputs the perception measurement score to the parameter control module. The parameter control module generates control commands based on the perception metric score and outputs them to the speaker protection processing module. The speaker protection processing module adjusts and updates the original audio signal according to the control command and then outputs the target audio signal.

2. The audio signal processing method according to claim 1, characterized in that, The perception measurement model includes a pre-trained psychoacoustic model; The perceptual measurement model performs perceptual difference analysis on the original audio signal and the first audio signal to obtain a perceptual measurement score, including: Using the pre-trained psychoacoustic model, the first auditory feature information of the original audio signal, the second auditory feature information of the first audio signal, and the distortion information between the original audio signal and the first audio signal are determined. Based on the first auditory feature information, the second auditory feature information, and the distortion information, the perceptual metric score is determined. The perceptual metric score includes auditory quality scores under different time dimensions.

3. The audio signal processing method according to claim 2, characterized in that, The perception measurement score includes a first time dimension score, a second time dimension score, and a third time dimension score; The first time dimension score reflects the instantaneous listening quality, the second time dimension score reflects the listening quality within a preset time period, and the third time dimension score reflects the long-term listening quality.

4. The audio signal processing method according to claim 3, characterized in that, The speaker protection processing module adjusts the original audio signal in real time to obtain the first audio signal, including: The speaker protection processing module inputs the original audio signal into the displacement protection model; The displacement protection model estimates the diaphragm displacement of the loudspeaker based on the original audio signal and outputs a first displacement protection parameter. The speaker protection processing module adjusts the original audio signal in real time based on the first displacement protection parameter to obtain the first audio signal.

5. The audio signal processing method according to claim 4, characterized in that, The control command includes a second displacement protection parameter; The parameter control module generates control commands based on the perception metric score and outputs them to the speaker protection processing module, including: The parameter control module adjusts the first displacement protection parameter based on the first time dimension score and / or the second time dimension score to obtain the second displacement protection parameter; The parameter control module adjusts the model parameters of the displacement protection model based on the third time dimension score. The parameter control module outputs the second displacement protection parameter and the model parameters of the displacement protection model to the speaker protection processing module.

6. The audio signal processing method according to claim 3, characterized in that, The speaker protection processing module adjusts the original audio signal in real time to obtain the first audio signal, including: The speaker protection processing module inputs the original audio signal into the temperature protection model; The temperature protection model estimates the voice coil temperature of the speaker based on the original audio signal and outputs a first temperature protection parameter. The speaker protection processing module adjusts the original audio signal in real time based on the first temperature protection parameter to obtain the first audio signal.

7. The audio signal processing method according to claim 6, characterized in that, The control command includes a second temperature protection parameter; The parameter control module generates control commands based on the perception metric score and outputs them to the speaker protection processing module, including: The parameter control module adjusts the first temperature protection parameter based on the second time dimension score to obtain the second temperature protection parameter; The parameter control module adjusts the model parameters of the temperature protection model based on the third time dimension score. The parameter control module outputs the second temperature protection parameter and the model parameters of the temperature protection model to the speaker protection processing module.

8. The audio signal processing method according to claim 5 or 7, characterized in that, The control command also includes a target pre-compensation gain parameter; The parameter control module generates control commands based on the perception metric score and outputs them to the speaker protection processing module, and also includes: The parameter control module adjusts the original pre-compensation gain parameters based on the third time dimension score to obtain the target pre-compensation gain parameters; The target pre-compensation gain parameter is used to adjust the global gain or the gain at a specific frequency point of the original audio signal.

9. An audio signal processing system, characterized in that, include: The speaker protection processing module is used to receive the original audio signal, adjust the original audio signal in real time to obtain a first audio signal, and output the first audio signal to the perception measurement model. The perception measurement model is used to receive the original audio signal and the first audio signal, perform perception difference analysis on the original audio signal and the first audio signal to obtain a perception measurement score, and output the perception measurement score to the parameter control module. The parameter control module is used to generate control commands based on the perception measurement score and output them to the speaker protection processing module; The speaker protection processing module is also used to adjust and update the original audio signal according to the control command and then output the target audio signal.

10. An electronic device, characterized in that, The system includes the audio signal processing system of claim 9 and a speaker, wherein the audio signal processing system and the speaker are electrically connected and are used to send the target audio signal to the speaker.

Images