Linear compensation method and audio system thereof

Abstract

The present invention provides a linear compensation method and an audio system thereof, which comprises the following steps: applying a test signal to a first sound-generating device; obtaining an audio measurement result generated by the first sound-generating device according to the test signal; generating a compensation curve according to the audio measurement result; and performing a linear compensation operation on a second sound-generating device according to the compensation curve. The present invention does not need to measure the point-to-point mechanical characteristics of the sound-generating device to find the correlation between electronics and mechanics, but can directly measure the audio characteristics of the sound-generating device corresponding to the input voltage signal.

Description

Technical Field

[0001] The present invention relates to a method for an audio system, and more particularly to a linear compensation method and an audio system thereof. Background Art

[0002] There is a linearity problem with sound producing devices (SPDs) such as speakers or headphones, where the sound producing device includes a diaphragm that can produce sound by being driven by an electronic signal (such as a voltage signal). Generally speaking, the voltage signal can drive the diaphragm of the sound producing device to move, and the mechanical displacement caused by this movement can compress air molecules to generate sound waves. The sound volume generated by the sound producing device has a nonlinear relationship with the input voltage signal, so linear compensation is needed to solve this problem.

[0003] The driving of the above-mentioned sound-generating device needs to be converted from an electronic signal into a mechanical displacement and then into an audio signal (i.e., a sound wave). In the known technology, a linear compensation curve is generated by executing a linear compensation algorithm and directly measuring the mechanical characteristics of the sound-generating device. There are many complex variation factors, such as the characteristics of the actuator used to drive the speaker, the mechanical design of the diaphragm in the sound-generating device, and the pressure formed during the manufacturing process of the sound-generating device. For example, for the most common piezoelectric material (PZT) actuator, its dielectric constant will drop rapidly as the voltage across the actuator increases. The piezoelectric material itself also has a certain degree of hysteresis, which may affect the displacement of the diaphragm according to the increase or decrease of the voltage. In addition, due to the change of hysteresis behavior, when the piezoelectric material actuator is in a dormant state and the applied voltage drops to 0V, the piezoelectric material will go to sleep, and it is necessary to use a "wake-up" procedure to restore the piezoelectric material to a normal working state.

[0004] The pressure design of the diaphragm (especially the pattern of the slits, that is, the thin lines cut through the thickness of the diaphragm to improve the consistency of the diaphragm) will greatly affect the displacement of the diaphragm. For example, the slit patterns and changes produced in the manufacturing process of the Micro-Electromechanical System (MEMS) may cause different pressures at different positions of the diaphragm, thus presenting different degrees of linear distortion. In addition, depending on the different signals applied during the measurement process, various diaphragm resonance modes may also cause displacement differences at different positions. Therefore, under the above-mentioned complex problems, it is difficult to measure at a limited number of measurement points in the diaphragm to obtain the correlation between mechanical measurements and audio results, making it impossible to establish a linear compensation curve with high enough accuracy for audio applications for the measurement of diaphragm displacement. In view of this, the known technology really needs to be improved. Summary of the invention

[0005] Therefore, the main purpose of the present invention is to provide a linear compensation method and a related audio system to solve the above problems.

[0006] An embodiment of the present invention discloses a linear compensation method for a sound producing device (SPD), the linear compensation method comprising the following steps: applying a test signal to a first sound producing device; obtaining an audio measurement result generated by the first sound producing device according to the test signal; generating a compensation curve according to the audio measurement result; and performing a linear compensation operation on a second sound producing device according to the compensation curve.

[0007] Another embodiment of the present invention discloses a linear compensation method for a sound-emitting device, the linear compensation method comprising the following steps: generating a sensitivity curve for a first sound-emitting device, the sensitivity curve comprising a plurality of sensitivity values; integrating the plurality of sensitivity values ​​to generate a plurality of linearity data; and generating a plurality of compensation data based on the plurality of linearity data.

[0008] Another embodiment of the present invention discloses an audio system for driving a first sound-generating device, the audio system comprising a memory and a computing circuit. The memory is used to store a plurality of compensation data. The computing circuit is coupled to the memory, and is used to receive a driving voltage for the first sound-generating device, and calculate a compensation voltage according to the driving voltage and a compensation data corresponding to the driving voltage among the plurality of compensation data. The plurality of compensation data are included in a compensation curve generated according to an audio measurement result, and the audio measurement result is generated by a second sound-generating device by applying a test signal to the second sound-generating device.

[0009] The present invention does not need to measure the point-to-point mechanical characteristics of the sound-generating device to find the correlation between electronics and mechanics, but can directly measure the audio characteristics of the sound-generating device corresponding to the input voltage signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 The figure is a flow chart of a linear compensation process according to an embodiment of the present invention.

[0011] Figure 2 The waveforms of a sensitivity curve and a linearity curve measured by the embodiment of the present invention are shown.

[0012] Figure 3 The figure is a flow chart of an audio measurement process according to an embodiment of the present invention.

[0013] Figure 4 A comparison of distortion before and after linear compensation is shown.

[0014] Figure 5 FIG. 1 is a schematic diagram of an audio system according to an embodiment of the present invention.

[0015] Figure 6 FIG. 4 is a schematic diagram of another audio system according to an embodiment of the present invention.

[0016] Figure 7 The figure is a flow chart of a process of an embodiment of the present invention.

[0017] Figure Number:

[0018] 10 Linear Compensation Process

[0019] 100~110、300~312、700~710 Steps

[0020] S-sensitivity curve

[0021] L Linearity Curve

[0022] N1 Node

[0023] 30 Audio Measurement Process

[0024] 50, 60 Audio System

[0025] 502 Memory

[0026] 504 Operational Circuit

[0027] 506 Power Amplifier

[0028] Cv linear compensation curve

[0029] V S Driving voltage

[0030] V LC Linear compensation voltage

[0031] V C Compensation voltage

[0032] 600 Feedback Circuit

[0033] 602 Low-pass filter

[0034] 604 Gain Amplifier

[0035] 606 Adder

[0036] 70 Process DETAILED DESCRIPTION

[0037] The present invention provides a linear compensation method to obtain a linear compensation curve according to a new audio measurement procedure, which can avoid the complex interactions between various electronic, mechanical and structural factors caused by the displacement of the brake diaphragm in a micro-electromechanical system (MEMS) sound producing device (SPD), and use the audio measurement data of the MEMS sound producing device to directly obtain the overall audio performance of the MEMS sound producing device to form a linear compensation curve.

[0038] Please refer to Figure 1 , Figure 1 FIG. 1 is a flow chart of a linear compensation process 10 according to an embodiment of the present invention. The linear compensation process 10 may be used to perform linear compensation on a sound generating device to generate a linear compensation curve. Figure 1 As shown, the linear compensation process 10 includes the following steps:

[0039] Step 100: Start.

[0040] Step 102: Generate a sensitivity curve for a sound-generating device, where the sensitivity curve includes a plurality of sensitivity values.

[0041] Step 104: Integrate the multiple sensitivity values ​​to generate multiple linearity data.

[0042] Step 106: Generate a plurality of compensation data according to the plurality of linearity data.

[0043] Step 108: Generate a linear compensation curve having a plurality of compensation data.

[0044] Step 110: End.

[0045] According to the linear compensation process 10, a sensitivity curve including a plurality of sensitivity values ​​can be generated for the sound-generating device (step 102). The sound-generating device can be, for example, a micro-electromechanical system sound-generating device such as the piezoelectric actuated device described above, which can be driven by a voltage signal to generate sound. Instead of measuring the point-to-point mechanical characteristics of the sound-generating device to find the correlation between electronics and mechanics, the present invention can directly measure the audio characteristics of the sound-generating device in response to the input voltage signal, wherein the audio measurement result can be represented as a sensitivity curve as in the linear compensation process 10.

[0046] In one embodiment, with respect to audio measurement, the sensitivity value on the sensitivity curve may be a sound pressure level (SPL), which corresponds to a test signal with a specific frequency and AC amplitude. Figure 2FIG. 2 shows the waveforms of a sensitivity curve S and a linearity curve L measured by an embodiment of the present invention. Figure 2 As shown, assuming that the operating voltage range of the sound-generating device is between 1V and 29V, the sound pressure level can be measured at a frequency of 300Hz and an input AC signal amplitude of 1V, with a bias voltage from 1V to 29V to generate sensitivity values ​​corresponding to 1V to 29V, thereby generating a sensitivity curve S. The measurement results show that the sound-generating device has the highest output sound pressure level when the input bias voltage is around 8V to 9V, which means that the sound-generating device is most sensitive when the input bias voltage is around 8V to 9V, and the sensitivity value gradually decreases as the input voltage increases or decreases.

[0047] Figure 2 The sensitivity curve S shown can be obtained by applying a test signal to the sound-generating device. The test signal can include a series of input voltage signals that can be used to perform audio measurements, such as a 300 Hz signal with a 1 V amplitude and a bias voltage between 1 V and 29 V. The sound-generating device can detect its sound pressure level while receiving the test signal.

[0048] Please refer to Figure 3 , Figure 3 FIG. 3 is a flow chart of an audio measurement process 30 according to an embodiment of the present invention. The audio measurement process 30 may be used to provide a test signal to generate a test signal such as Figure 2 The sensitivity curve S shown in FIG. 1 can obtain the audio characteristics of the sound-generating device within its operating voltage range of 1V to 29V. Figure 3 As shown, the audio measurement process 30 includes the following steps:

[0049] Step 300: Start.

[0050] Step 302: Execute a wake-up procedure.

[0051] Step 304: Apply a slow large signal driving voltage.

[0052] Step 306: Mix the slow large signal driving voltage with a small signal.

[0053] Step 308: Place the sound-generating device in a measurement environment.

[0054] Step 310: Measure the sound pressure level at the frequency of the small signal and record the sound pressure level as the sensitivity value.

[0055] Step 312: End.

[0056] According to the audio measurement process 30, a wake-up procedure is first performed. In the wake-up procedure, a wake-up signal can be applied to the sound-generating device. The wake-up signal can be a 1kHz sine wave signal oscillating between 1V and 29V. The wake-up signal can drive the sound-generating device into its normal operating state. Then, a slow large signal driving voltage and a small signal can be generated and mixed with each other. In one example, the large signal driving voltage can be a 3Hz sine wave signal oscillating between 1V and 29V, and the small signal can be a sine wave signal with a frequency of 300Hz, and its amplitude is equal to 0.5V or 1V, but is not limited thereto. Then, the sound-generating device is placed in a measurement environment, such as an ear emulator. The ear emulator can measure the sound pressure level of the sound-generating device at the frequency of the small signal (such as 300Hz), and record the sound pressure level at the target voltage value (from 1V to 29V) as the sensitivity value. Since the small signal is carried by a slow large signal oscillating between 1V and 29V, the sound pressure level at the small signal frequency can be automatically measured when the large signal changes to the target voltage value, thereby generating a sensitivity curve S from 1V to 29V.

[0057] It is worth noting that the above steps are only examples for illustration, and the details of each step or the order or combination of the steps may vary depending on the corresponding device to be tested and the measurement target or result. For example, the wake-up procedure (step 302) may be repeated during the measurement step (step 310). In addition, the frequency and / or amplitude of the large signal and / or the small signal may be other appropriate values, or vary between a set of different values ​​rather than fixed values. In addition, the large signal may also oscillate for N cycles (preferably N is greater than 50), and the sound pressure levels measured during these cycles may be averaged to obtain sensitivity data.

[0058] Alternatively, the small signal may be made to oscillate at a plurality of target voltage values ​​to measure the sound pressure level, that is, the voltage value may be manually changed to replace the slow large signal to achieve the target voltage values.

[0059] Please refer back Figure 1 and Figure 2 After obtaining the sensitivity curve S of the sound generating device, the sensitivity curve S can be used to generate the linearity curve L. For example, the sensitivity values ​​can be integrated or summed to generate linearity data, thereby forming the linearity curve L (step 104). The integration operation can be expressed as follows:

[0060]

[0061] After the linearity curve L and the linearity data of the sound generating device are determined, compensation data for the sound generating device can be generated according to the linearity data (step 106). In one embodiment, the compensation data for the sound generating device can be obtained by calculating the inverse of the linearity data, and then a linear compensation curve Cv with the compensation data is generated (step 108). The linearity curve L and the linear compensation curve Cv satisfy the following relationship:

[0062] L×Cv=Const; (2)

[0063] The value Const may be 1 or other suitable constants.

[0064] From another perspective, the linearity curve L and the linear compensation curve Cv can be regarded as a mapping relationship or transfer function, and L×Cv can be regarded as a composite transfer function of L and Cv. Equation (2) can be interpreted as the slope of the composite transfer function L×Cv should be a constant (i.e., a straight line) or a linearized curve.

[0065] That is to say, for an (ideal) linear sound-generating device, L should have the following characteristics: L(Vin) = a·Vin, where a is a positive constant (such as a=1), and L(·) can be regarded as a function (mapping relationship) of mapping the (input) electronic signal Vin to the mechanical diaphragm movement / displacement L. In practice, the sound-generating device is usually nonlinear, that is, L(Vin)≠a·Vin, so a compensation curve / operation Cv is needed to improve the overall linearity so that Cv(L(Vin)) is a straight line with respect to Vin, which can be expressed as Cv(L(Vin)) = a'·Vin, or any other linearity / linearization curve (such as a curve or mapping relationship corresponding to Dynamic Range Compression / Compensation (DRC)). It should be noted that the constant a' may be the same as or different from the constant a, and Cv(·) represents a mapping relationship.

[0066] It can be seen that in the linear compensation process 10 and its related audio measurement process 30, the linearity curve L and the linear compensation curve Cv of the sound-generating device are calculated and obtained by the sound pressure level generated by the sound-generating device, wherein the measurement of the sound pressure level has the characteristics of pure audio, and does not need to consider the diaphragm displacement or any other mechanical characteristics of the sound-generating device. Therefore, the data used for linear compensation is only related to the audio phenomenon of the sound-generating device, and does not need to measure the mechanical characteristics of the sound-generating device, such as the displacement of the diaphragm. In this way, it can be free from the influence of various factors that complicate the correlation between mechanical characteristics and audio performance and make mechanical measurement difficult.

[0067] Please refer to Figure 4 , Figure 4The distortion comparison of an actual MEMS sound generator before and after linear compensation is shown. Figure 4 The total harmonic distortion (THD) measured on a frequency band by the linear compensation method proposed by the present invention is shown, and the total harmonic distortion without linear compensation is shown for comparison. Figure 4 As shown, linear compensation can achieve a nonlinearity reduction rate of more than 90% in the frequency band. For example, in the frequency band of 100Hz to 200Hz, the total harmonic distortion before linear compensation is approximately between 5.6% and 6.2%; if the linear compensation of the present invention is performed, the total harmonic distortion will be reduced to a range between 0.3% and 0.5%, which is equivalent to a significant improvement of 92% to 95%.

[0068] Please refer to Figure 5 , Figure 5 FIG. 5 is a schematic diagram of an audio system 50 according to an embodiment of the present invention. Figure 5 As shown, the audio system 50 includes a memory 502, a computing circuit 504 and a power amplifier 506. The audio system 50 can receive a driving voltage V S According to the above linear compensation method, the driving voltage V S Compensation is performed to generate a linear compensation voltage V LC , and the linear compensation voltage V LC In other words, the audio system 50 can perform a linear compensation operation on the sound-generating device according to the linear compensation curve Cv.

[0069] Specifically, the memory 502 may be used to store a linear compensation curve Cv, which includes a linear compensation curve corresponding to the driving voltage V S Compensation data, V S The value of is within the operating voltage range of the sound-generating device. The linear compensation curve Cv and the compensation data can be obtained through the above linear compensation process, that is, the linear compensation curve Cv can be generated by calculating the inverse of the linearity curve L, and the linearity curve L can be generated by integrating the sensitivity curve S, and the sensitivity curve S is the result of audio measurement of the sound-generating device.

[0070] The operation circuit 504 can be configured to generate a voltage V S and the corresponding driving voltage V obtained from the memory 502 S The compensation data generates a compensation voltage V C In one embodiment, the operation circuit 504 may perform a curve fitting operation to calculate the compensation voltage V CThe computing circuit 504 may be implemented in, for example, an application specific integrated circuit (ASIC) or a processing circuit with computing capabilities.

[0071] In one embodiment, the operation circuit 504 may include a multiplier. The multiplier in the operation circuit 504 may receive the driving voltage V S , and the driving voltage V S Multiply by the driving voltage V S The compensation data obtained / calculated is used to adjust the driving voltage V S Compensation is performed to generate a compensation voltage V C . The compensation voltage V C With improved linearity.

[0072] When the compensation voltage V C After generation, the power amplifier 506 may receive the compensation voltage V C And output corresponding to the compensation voltage V C Linear compensation voltage V LC The power amplifier 506 can be used to provide sufficient driving capability to drive the load of the sound-generating device. In one embodiment, the linear compensation voltage V LC Can be an analog voltage, and the compensation voltage V C In one embodiment, from the perspective of linear compensation, the power amplifier 506 can be selectively set or designed to have a unity gain. In this case, the compensation voltage V C Can be regarded as equivalent to or equal to the linear compensation voltage V LC .

[0073] The linear compensation operation provided by the audio system 50 enables the following conversion:

[0074] V LC =V S ×Cv. (3)

[0075] It should be noted that the purpose of the power amplifier 506 is to improve the driving capability, and its gain is equal to 1.

[0076] It should also be noted that the measurement of the sensitivity curve S and the calculation and generation of the linear compensation curve Cv can be performed before the sound device product leaves the factory, and the obtained linear compensation curve Cv is then stored in the memory 502. In one embodiment, each sound device product should perform audio measurement to obtain the sensitivity curve S and generate the linear compensation curve Cv accordingly, and the linear compensation operation of this sound device can be performed according to its corresponding linear compensation curve Cv. For example, an audio measurement can be performed on a first sound device to generate a sensitivity curve S, and its related linear compensation curve Cv can be used for compensation of the first sound device. In another embodiment, one or more samples can be selected from a batch of sound device products to measure the sensitivity curve S and generate the linear compensation curve Cv accordingly, which can be used for compensation of the batch of sound device products. In this case, an audio measurement can be performed on a first sound device to generate a sensitivity curve S, and its related linear compensation curve Cv is used for compensation of a second sound device, which is different from the first sound device.

[0077] In the audio system 50, the power amplifier 506 may have AC coupling at the feedback stage or the input analog signal may be AC ​​coupled to its input stage. Therefore, due to the linear compensation imbalance characteristic presented in equation (3), voltage drift may occur. For example, Figure 2 As shown in the sensitivity curve S, the sound device is most sensitive around 8V to 9V, rather than at the middle voltage of the operating voltage range of 15V. The shift of this sensitivity peak causes the center point of the device linearity curve L to be mapped to the voltage level of 13.3V. Figure 2 As shown, the corresponding dotted line intersects at node N1, indicating that there is a voltage drift of -1.7 V from the intermediate voltage level (1+29) / 2=15 V. If this voltage drift is not compensated, the linear compensation of equation (3) will be invalid, because of the mismatch between the actual device linearity and the linearity curve model.

[0078] The voltage drift of the linear compensation curve can be compensated in the audio system by using a feedback circuit. Figure 6 , Figure 6 FIG. 6 is a schematic diagram of another audio system 60 according to an embodiment of the present invention. Figure 6 As shown, the structure of the audio system 60 is similar to that of the audio system 50, so signals or components with similar functions are represented by the same symbols. The difference between the audio system 60 and the audio system 50 is that the audio system 60 further includes a feedback circuit 600 for compensating for voltage drift. The feedback circuit 600 may include a low pass filter 602, a gain amplifier 604 and an adder 606.

[0079] The low pass filter 602 can be used to output the frequency of the AC coupling effect that causes the voltage drift. Therefore, it is necessary to modify equation (3) to V LC =(V S -V D.LC )×Cv to perform matching level shift, making AC coupling invalid, thereby correcting the linearity curve model to be consistent with the linearity of the actual device.

[0080] However, the new factor V D.LC A negative feedback loop may be formed and cause oscillation at low frequency, so appropriate stability analysis of gain margin and phase margin is required to ensure that low frequency oscillation does not occur. The gain amplifier 604 and adder 606 can be used to achieve this purpose. The equation can be modified as follows to improve stability:

[0081] V LC =(V S -(k·V D.AC +V D.0AC ))×Cv; (4)

[0082] Among them, V D.0AC To correct the midpoint of the device linearity to zero voltage, for example, the expected -1.7V drift, as Figure 2 As shown, the factor k·V D.AC The signal output from the low-pass filter 602 is multiplied by k through the gain amplifier 604 to compensate for the AC coupling signal and ensure loop stability.

[0083] The above linear compensation process and the operation of the audio system can be summarized into a process 70, such as Figure 7 The process 70 includes the following steps:

[0084] Step 700: Start.

[0085] Step 702: Apply a test signal to a first sound generating device.

[0086] Step 704: Obtain an audio measurement result generated by the first sound generating device according to the test signal.

[0087] Step 706: Generate a compensation curve according to the audio measurement result.

[0088] Step 708: Perform a linear compensation operation on a second sound generating device according to the compensation curve.

[0089] Step 710: End.

[0090] In process 70, the second sound-emitting device may be the same as or different from the first sound-emitting device. For other detailed implementations and variations of process 70, reference may be made to the description in the above paragraphs and will not be repeated here.

[0091] In summary, the present invention provides a linear compensation method for an audio system. Instead of measuring the mechanical characteristics of a sound-generating device to find out the correlation between electronics and mechanics, the present invention can directly measure the audio characteristics of the sound-generating device in response to an input voltage signal. Therefore, the data used for linear compensation is only related to the audio performance of the sound-generating device. In this case, it is free from the influence of various factors that complicate the correlation between mechanical characteristics and audio performance and make mechanical measurement difficult. The linear compensation method proposed by the present invention can achieve a nonlinearity reduction rate of more than 90% in important frequency bands.

[0092] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A linear compensation method for a sound generating device, characterized in that: include: Applying a test signal to a first sound generating device; Obtaining an audio measurement result generated by the first sound generating device according to the test signal; generating a compensation curve according to the audio measurement result; as well as According to the compensation curve, a linear compensation operation is performed on a second sound generating device to compensate for a linearity of the second sound generating device with respect to a driving voltage for the second sound generating device.

2. The linear compensation method according to claim 1, characterized in that: The step of obtaining the audio measurement result generated by the first sound generating device according to the test signal comprises: When the first sound emitting device receives the test signal, a sound pressure level of the first sound emitting device is detected.

3. The linear compensation method according to claim 1, characterized in that: The test signal includes a sine wave signal carried at a plurality of voltages.

4. The linear compensation method according to claim 3, characterized in that: According to the audio measurement result, the step of generating the compensation curve comprises: measuring a plurality of sound pressure levels of the first sound emitting device corresponding to the plurality of voltages; integrating the plurality of sound pressure levels to generate a plurality of linearity data; and generating a plurality of compensation data according to the plurality of linearity data; The plurality of compensation data are used to establish the compensation curve.

5. The linear compensation method according to claim 4, characterized in that: According to the compensation curve, the step of performing the linear compensation operation on the second sound generating device includes: multiplying the driving voltage for the second sound generating device by a compensation data corresponding to the driving voltage among the plurality of compensation data to generate a linear compensation voltage; and The linear compensation voltage is output to drive the second sound generating device.

6. The linear compensation method according to claim 4, characterized in that: According to the plurality of linearity data, the step of generating the plurality of compensation data comprises: The reciprocals of the plurality of linearity data are calculated to generate the plurality of compensation data.

7. The linear compensation method according to claim 1, characterized in that: The compensation curve is obtained without measuring the mechanical characteristics of the first sound generating device.

8. A linear compensation method for a sound generating device, characterized in that: include: Generating a sensitivity curve for a first sound-generating device, the sensitivity curve comprising a plurality of sensitivity values; Integrating the multiple sensitivity values ​​to generate multiple linearity data; as well as generating a plurality of compensation data according to the plurality of linearity data; The plurality of compensation data are used to compensate for a linearity of a second sound generating device relative to a driving voltage for the second sound generating device.

9. The linear compensation method according to claim 8, characterized in that: According to the plurality of linearity data, the step of generating the plurality of compensation data comprises: The reciprocals of the plurality of linearity data are calculated to generate the plurality of compensation data.

10. The linear compensation method according to claim 8, characterized in that: Also includes: A linear compensation operation is performed on the second sound generating device according to a compensation curve composed of the plurality of compensation data.

11. The linear compensation method according to claim 10, characterized in that: The step of performing the linear compensation operation on the second sound generating device according to the compensation curve composed of the plurality of compensation data comprises: multiplying the driving voltage for the second sound generating device by a compensation data corresponding to the driving voltage among the plurality of compensation data to generate a linear compensation voltage; and The linear compensation voltage is output to drive the second sound generating device.

12. The linear compensation method according to claim 8, characterized in that: The plurality of sensitivity values ​​include a plurality of sound pressure levels of the first sound emitting device under a test signal.

13. The linear compensation method according to claim 8, characterized in that: The plurality of compensation data are obtained without measuring the mechanical characteristics of the first sound generating device.

14. An audio system for driving a first sound-generating device, characterized in that: The audio system includes: a memory for storing a plurality of compensation data; and an operation circuit coupled to the memory, for receiving a driving voltage for the first sound-generating device, and calculating a compensation voltage according to the driving voltage and a compensation data corresponding to the driving voltage among the plurality of compensation data; wherein the plurality of compensation data are included in a compensation curve generated according to an audio measurement result; The audio measurement result is generated by a second sound-generating device by applying a test signal to the second sound-generating device; The plurality of compensation data are used to compensate for a linearity of the first sound generating device relative to the driving voltage for the first sound generating device.

15. The audio system of claim 14, wherein: Also includes: A power amplifier is coupled to the operation circuit and the first sound generating device, and is used for outputting a linear compensation voltage corresponding to the compensation voltage to drive the first sound generating device.

16. The audio system of claim 14, wherein: The audio measurement result includes a sound pressure level of the second sound emitting device.

17. The audio system of claim 14, wherein: The compensation curve is generated without measuring the mechanical properties of the second sound generating device.

18. The audio system of claim 14, wherein: The operation circuit includes a multiplier, and the multiplier is used for multiplying the driving voltage by the compensation data.

19. The audio system of claim 14, wherein: The compensation curve is generated by performing the following steps: measuring a plurality of sound pressure levels of the second sound emitting device corresponding to a plurality of voltages; Integrating the plurality of sound pressure levels to generate a plurality of linearity data; as well as The plurality of compensation data on the compensation curve is generated.

20. The audio system of claim 19, wherein: The compensation curve is also generated by performing the following steps: The reciprocals of the plurality of linearity data are calculated to generate the plurality of compensation data on the compensation curve.

21. The audio system of claim 14, wherein: Also includes: A feedback circuit is used to compensate a voltage drift of the compensation curve.

22. The audio system of claim 21, wherein: The feedback circuit includes: a low pass filter for outputting a frequency corresponding to the voltage drift; and A gain amplifier and an adder are coupled to the low pass filter to improve the stability of the audio system and compensate for the voltage drift.

23. The audio system of claim 21, wherein: The voltage drift comes from an AC coupling in the audio system.

Images