TW479437B - Compensation system and method for sound reproduction - Google Patents

Abstract

A sound compensation system alters an electrical audio signal for input to a sonic reproduction device having associated behavioral characteristics. The behavioral characteristics of the device are defined by individual or groups of individual components of the sonic reproduction device and include mechanical, acoustic and electromagnetic behaviors. The model includes a plurality of filters that simulate at least one of the behavioral characteristics of the sonic reproduction device. The filters are defined by digital signal processes or by analog circuits and are characterized by one or more of an associated frequency, time, phase and transient response. These responses combine to define an overall response for the model. The filters include adjustable parameters which are used to alter filter responses to produce responses that are conjugates to the responses of the unaltered filters and thus the sonic reproduction device. A controller modifies the parameters.

Description

1 ~ M Vl > l BACKGROUND OF THE INVENTION Field of the Invention: Compensator i = is related to a kind of interpolated music system used in the complement of audio transmission and reproduction system, and particularly relates to a method of using parameter values to control or adjust Cheng Cheng =! Method and system 'and the program has a knife or 70 piece conversion of the transmission and regeneration system, the nature of the model, or the response. Known technical description: Lida's audio reproduction system uses electromechanical speakers to make acoustic sounds. The electronic, mechanical, and auditory properties of the speaker are often worse than the heart's n.¾, and can cause distortion, irregular response, and other color noise. Many techniques have been used to compensate the characteristics of the speaker to improve the hearing texture. ^ The function or behavior model of the speaker has been used in practice, and it has been developed on the basis of complements. "Act ive Equal izat ion of Loudspeakers", Speaker Builder, February 1997 is an example of a model ', which illustrates the procedure of the model. These models combine technical language and are usually stimulated by electronics to mimic or simulate the auditory response of a speaker system. The generation or synthesis of a model is usually a function of a group of components, which is the overall performance or behavior of the speaker. Lines & magnets (pads) form a motor, which is represented by resistors, inductors, capacitors, back-EMF generators and other converted components. Air loudness, moving mass, acoustic loading, magnetic-braking, and mechanical loss factors can be divided by the LRC resonator network or circuit.

1012-3382-PF.ptd Page 4 I Circle

479437

Analysis and simplification. These electromechanical, auditory, and mechanical representations converted in the model can be simplified or reduced to less than 30 components. The model's response still resembles that of a horn, but these gaskets make the model no longer have the speaker element properly 仃 equivalent. Therefore, the traditional model can neither predict nor use zero phase compensation when using the speaker. , ’/ 'We can characterize an ideal imaginary (weU-conceived) model as well as other properties, and correct the linear phase of the loudspeaker. This technology works in a form, but because it is subject to a dedicated, rigid circuit or a special program of the traditional model, it can only be designed as a single speaker due to the restrictions. Some high-quality crossover online books divide the signal spectrum of multiple driven signals, and the spectrum shows a conjugate response similar to the above. 'A low-frequency resonant boost (l0w-frequency resonant b〇st) was deliberately designed in the speaker. Traditional models often cannot measure and predict auditory behavior. This method is similar to that of other ::: Counting Parameter One | 'will affect the stress response. The air :: sex determines the frequency, resonance loss, and structural efficiency of the auditory output of the speaker. In the wide frequency range, it looks like a zero-phase conjugate or the same order correction can be designed and presented in the ^ 敕 t method. Several components need to match this resonance characteristic, but at the% yield it is necessary to determine all internal interactions. / = Ground control, for example, when adjusting it from "= = true Π = and correction, Ba Ba system must be modified according to experiments or

Need to make constant adjustments when analyzing the system ... The lumped model is set to S

479437 V. Description of invention (3) A specific Yang design. Traditional materials, when set to meet the requirements, the method requires a correction of the phase response. If the phase combination will interfere with some of the Yang energy, the patterns will affect or break each other. Signals can be generated at other frequencies, or any phase due to sound can be difficult to get from the traditional spectrum. For example, wave behavior. Affects these responsive microphones. For other speakers, the procedure needs to be re- = CurVe-fitting requires hundreds of resources ^ So the corresponding adjustments, and the need for permission? Component or power supply II frequency response. The analog method is impractical, and the digital operation and the extensive structure. They can neither provide accuracy, nor can they be implicitly determined without speaker data or operation. Without a model, the amplitude, time and sum obtained from the measurement become very scary. The important behaviors of the horn (with effects on the auditory senses) are not or achieved by traditional methods. Such behaviors include standing wave ruptures and coupled resonances, and non-linear results may affect auditory and mechanical behavior. Random random months with adverse consequences occur. Even when the average response remains flat or consistent across its 8 guards, the energy will increase during the release period when the signal changes or ends. In addition, others related to the mobile unit ^ and The spatial factors of high-frequency decoupling removed by the loop also need to be considered. This number will cause signal source movement, delayed energy release, and three-dimensional errors. Such a damage response can be invisible or very sleepy microphone-spectrum analyzer ( microphone-and-analyzer) correction method. Say that nodal will produce unwanted response, and the stationary time, directional behavior, and light output of the resident speaker. Signals that often cause invisible perception or cognition are Changed

V. Description of the invention (4)

The mechanical movement has a large number of U pairs; the phase difference of the pieces. Auditory d: Excluded: The position where the transducer can determine the signal action is flat, but humans reduce the imaging by 1 body sound. And the change of 'human perception can be mechanically perturbed. It is often heard to hear 3 from measuring the response, or to change with the frequency sweeper and bit relationship: the radiation surface can be stored in different phases. Move low. When the sense of rotation is better than the transducer and the energy, the signal can be the same! C suddenly stops releasing. The beating sound between two frequencies can be heard, and the rates affect each other. This kind of sound in the frequency range will become louder and louder ^ very annoying. Falling on the interference frequency, the perturbation bit of the resolution driver is granular. Human two-ear hearing devices can provide more frequencies for the stereo reproduction of multiple speakers: shall = damage j: for this case, mechanical response. Experiments have shown that the use of chrome and mutilation to protect the inevitable response during regeneration. The energy removed to form a deep trap can set the correction frequency to a flat output.耒 said "/, you can make an unwanted balance loss method = type: Γ device is composed of a transducer, protection and some resonance parts usually extend the bass response in the actual and through or controlled by the transmitter. This rate responds to the accuracy of 'bass extension and acceptable distortion. F = J, = higher efficiency while reducing amplification related to power supplies and packaging. This compromise will make many unwanted actions

Page 7 479437

V. Description of the invention (5) is manifested. Most traditional speaker correction methods use changes in the amplitude equalizer to flatten and extend the response of the speaker. Sometimes use your ears to make adjustments. For quantification, appropriate information is needed. Most common techniques use a spectrum to analyze the noise stimulus, and then respond to the day to indicate how to adjust the equalizer. More old technologies are based on the delay test or sampling window, the first muting from the loudspeaker-the measurement is based on the length or its delay group and the combination is determined, but the scale, the wave energy caused by the need to be put into the dimension, the environment It is necessary to set team rules. Quotient 7 — The decision should be taken to take measurements and remove disturbances to the two houses to produce data. The purpose is to obtain appropriate information from a listener or practical standards. More known systems use multiple delays to adjust the path turn-on delay to line up multiple speakers. His technology provides transient response waves, waterfalls, or continuous spectrograms. Multiple time-related messages are required. Such information needs to be explained, and the rate of use should be limited. Some behavioral responses must be spoken, and the information of the two speakers must be known. Measuring equipment = plus; a differentiating auditory probe like a microphone is required. The instrument can ^ =; Ϊ: near the cymbal, and move to detect the change in position i; k. ▲ There is subjective observation noise generated Adjust or slowly stretch through the magnetized frequency Hearing position, such as floor, books •, electricity: for the daggers, we should have a small ::: second product. Specifications or representations of other technologies; phase-accuracy modified common vehicle models. Eight-necked 々;}: Father is thorny, manual 1M ^ J uses the adjustment of the graphic specializer to assume ^ t of the industry: ί; wave waver, _ crossover device design analogy-there can be restrictions To perform these functions. Graphics etc. 4

H / V437, invention description (6) bands or resonators, and parametric filters include §positive filters and a few variable crossover and separation functions (sheIf ti〇 ^) and require more filtering Device. That is, the = effect = compensation behavior in a speaker does not have the correct phase and time response. Combining the graph and the parameterizer cannot provide sufficient explanation. Any correction cannot match 5 pts: Qiaoqian, so there will be phase errors. • There are insufficient filters to set the time and standing wave problems. Group delay attenuation, time-phase error, incorrect correction, and other shortcomings may become other improvements. Digital signal processing filters can generate more filtering sections than actual analog circuits. The graphic equalizer uses the parameter control section to make up for more response steps. Such a procedure is difficult to establish because the interference of the internal = has significant identification of behavioral instructions. Without input, the conjugate response cannot be modified. Standing wave and knot attenuation can be corrected with this system. However, unskilled editing and programming f must be ruled out, otherwise it is difficult to compensate for notches. The most likely method is to collect a single point response and use fast Fourier transform (FFT) input to the system. Such methods cannot respond or provide time-phase information to the true conjugate response required by the speaker. Like the MLSSA analysis system, the internal interference can be removed, and the frequency, transient, and stable response of the speaker system can also be generated. However, a large amount of data is required to perform these measurements. Therefore, a multiple-band graphic equalizer is not a good choice for installing corrections. Digital signal processing systems can generate many parametric filters and related time programs made using specific circuits. Traditional large digital signal processing

479437 V. Description of the invention (7) The system can use the phenomenon of measurement to confirm and select the behavior of the speaker. Our frequency domain response is likely to add phase error and check the amount of delay stability. Sound can be improved by one listener position, but for others =

In terms of it will be reduced. Most likely, there is a change in the reproduced sound, and once there is no way to improve it. w ... I Therefore, the sound reproduction needs to consider the system's identification requirements and a model that can produce the complete behavior of the co-roller response of the Lixuan regeneration device. Modifiable covariance ^ Requirements and methods must also be considered. In addition, the compensated reproduced sound environment must also be identified. The present invention fulfills these and other requirements. ， 衣 发明 发明 简 简 Modeling into the equipment balance, there is a plurality of units. This includes up to a plurality of this modification to provide the adjustable level. The number of the system and the frequency of the system is clear, the sound is clear, the first type is the individual frequency of the device, and the frequency can be modified by the filter and the audio reproduction parameters. Each of the individual inventions can be adjusted. This invention proposes a method for establishing audio reproduction equipment and orientation, subtraction, and definition of a phase or device related to the sound device. The parameters of the audio signal and each method of filtering each are individually. In order to reduce the attenuation and reduce the attenuation, the audio frequency response time of the present invention should be combined with the response rate, time, and each adjustable integer of the audio regeneration electronic language device allows adjustment of the response of the yoke. The method produces a total amplitude and provides improved stereo sound. A modifiable electronic voice full response. Every transient response. Don't ring the dust, change the electronic number. The device at least one modifies the response of the wave filter to the modified conjugate yoke model and improves the sound imaging. The input to a signal device has a unique response. This setting includes changing the filter. Speech signals and more include complex filter phases. This adjustment should be related to

One less don't respond. The error is caused by the establishment of a complex oblique filter or network, tf, the response and the establishment of adjustments to allow the conjugate response. ^: F /, yoke filter or network, the system is quasi-responsive. 特殊 For special reaction Or modify the response of the playback device ί ::::: η, multiple detailed aspects of the audio playback device, such behaviors are ㈣. In the full change of the other playback device.;:, Put two & formation A full response conjugate to the microphone includes-cutoff filter, and at least-modified filter = parameters ㈣ frequency, amplitude and Q ° parameters ^. Cutoff ^ frequency response of the wave filter ΐ; t, the wave filter includes a constant slope Equalizer, one two two seven Another-in detail, at least one or two of the frequency and interval parameters. In the device, and adjust the parameter notch; ;; =-parameter notch filtering = sheep, amplitude and Q The parameters and the number include notch and notch lifting filters, and the modified filter includes a parameter number including the notch μ γΓ wave lift and the frequency response of the wave filter, notch and frequency, amplitude, and Q parameters. ., 翏 In the second form, this is sent: variable-electronic voice signal is input to a tone Qi :: Behavioral characteristics of the message compensation system. The system includes a tone 4 regeneration device with a related model including a plurality of simulations at least. The model includes a wavelet or program. Each filter cry and the message regeneration Full response of split money traits. Each response includes Heqing 1012-3382-FF-ptd

Page 11 5. Description of the invention (9) Time, phase or temporary departure. Xi ^ > Control of the waver response ί :; get ;, = change the filter for each multiple 遽 response. Each :: = Bo Yi can be converted into a kind of conjugate to the original or the sound of the program ;. The mother-to-vehicle filter has the same filter device as the original filter device :::::: J: Second class. The characteristic is to reproduce individual components and components or elements by audio. Hunting is based on the parameters of the model regeneration device. Establish the use of buckles. Because of the seven, the characteristics of these individual compensations have the same components or characteristics: these same compensations can be used in a logic device or by analogy; J :: JJ Circuit elements are adjusted by digital signals. In two aspects, and the controller includes a computer or speaker and at least one sentence cup, the audio regeneration device includes a waver, and the parameter value is adjusted by two kinds of adjustable parameters The complex filtering: in detail, 'equipped with this controller ;: in another device, monitor the setting of i sound ==. "-In detail, the parameter value of the control status. The program status and settings of the device are based on at least one program based on η: About -s audio system. The sound reproduction device. The audio including the audio behavior characteristics can output an electronic voice signal to the audio reproduction device model. At least one of the audio reproduction devices i :: bamboo type has multiple modifiable filters. Every _ _ &%, mechanical, auditory, and electromagnetic behavior characteristics. Considering the benefits have-including at least-a frequency, time, 479437 V. Description of the invention (10) related to the response. The model outputs an electronic voice signal to the regenerator. The audio system further includes a controller that can modify the plastic application, so that the filter convertible model becomes a kind of = :: conjugate model of the wavelet and the wavelet is lighter than the original wavelet. Among the four types, the present invention relates to a method for electronic voice signals of a message reproduction device. The == to-tone characteristic of a plurality of individual responses. These individuals have the full response of the audio reproduction device. The response of each individual; boundary ... seeder is such that for each filter, and the multiple wave responses are equal to one of the other responses. On this side, the positive response includes the steps of a tone signal to the filters. 'Enter the electronic language in a more detailed form of the invention, to one, f, and adjust the frequency of the cut-off filter: "-cut-off filter rate, amplitude and Q parameters. In another detail, β should The parameters include the peak frequency coefficient slope equalizer, and adjusting the filter to J. The filter includes parameters including the crossover frequency and the number of interleaved tables. The rate response equalizer includes a parameter notch filter. In another detailed aspect, the parameters of at least the frequency response include notch & adjusting the parameter of the notch filter is another detailed aspect, at least one of the amplitude, amplitude, and Q parameters. And adjust the parameter notch i. The f wave filter includes-the parameter notch includes notch frequency, amplitude and Q parameters. ~ / Parameter frequency response parameter package

479437

In the fifth form, the signal input method includes a defined rate and a time-based method. In addition, the filter includes a logical parameter. The parameter obtained by the method is used to generate an audio signal. Steps with plural traits. Each filter has a full response interval, phase, or transient response of the model. Each complex conjugate can be modified into a more detailed invention. The speaker includes at least one phase and the filter is modified to calculate at least From a standard speaker value, the actual adjustable parameters are used; in a detailed aspect, the steps of the method. In the other wave filter package, the tunable gains are empirical and the settings have related analog to wave filters with each response. For the original response of each filter, the audio level can be adjusted to include the following step parameter values; at least one method can be used to change the characteristics of an electronic voice. The response of the audio regeneration device is based on at least one frequency, and the method causes the number of responses to be set to at least one program condition; and a basic parameter. The standard test determines the value more including the modulation setting, a detailed aspect, the behavior corresponding to at least one, a filter corresponding to a filter response, a regenerative setting parameter, and a set of adjustment parameter measurement method parameters. At least one method monitoring program includes a plurality of numbers that have been counted in the speaker, and it is determined that the setting of the calculated value of a speaker filter in the invention parameters for the audiovisual condition is to take one less of the other parameter in one phase. Defining at least some special names and abilities to maintain sound neutrality can make a complex, tragic change response compensation system and method, and can be controlled and adjusted with simple and intuitive control specifications. Compared with the traditional response level (^ espfnse-leveling) method, the compensation system and method require less program complexity and can be applied to different audio reproduction

1012-3382-PF.ptd Page 14 479437

system. The compensation system and method can be turned on or off, shifted from one frequency to another, or other simple, intuitive changes. Combining the formulas of: f 其他 The system of the present invention and other modes and advantages will be more clearly and better represented by the following embodiments. Brief Description of the Drawings Figure 1 is a block diagram showing an audio reproduction device including a compensation system. Fig. 2 is a block diagram showing a compensation system. The system includes: each modification filter includes multiple slaves of frequency, time, and phase response. Figures 3a to 3e show the response diagrams of multiple components of a loudspeaker. Figure 3J shows the response diagrams in combination with the figures 3 to 3. 4 The Wishfulness of the Stomach Chapter 4 a Picture $ 繁 4 p 区! β modified response; every y 糸 shows the response of multiple adjustable elements of a compensation system. θ a can be adjusted to a total of lighter than the responses from Figures & 3e to Figure 4e. Figure 4f shows the total response of the company's car. The individual responses of individual frequencies in Figures 4a @ to 4e are shown in Figure 5. This figure shows the manufacturing of a vertical cone. Figure 6 shows the display. Figure 7 shows the frequency response of a small speaker driver with an acoustic watt wheel. Exploded map, in which the arrow on the right shows the increase / V. Description of the invention (14) Adjusting the parameter 22 Detailed description of the preferred embodiment: The above description will focus on the application of the present invention to a speaker system. The "invention" invention is not limited to such an application, but can be applied to a direct-sense transmission and reproduction device. Eight-Now referring to the figure, where the reference numerals in the figure represent the relative =: 'as shown in Figure 1-corresponding to the system of the present invention. The system package provides a signal source 12 for an electronic voice signal. For example, the signal is a CD player. The output 14 of the signal source 12 is the input of a compensation system 16. The compensation system 6 describes the invention in detail. The "output" is processed by the compensation system 16 and is generated-CD player compensation = Chuzi Voice # No. 18, the electronic voice signal 18 is input to a CD player power amplifier 19 and a CD player speaker 20. Referring to Figure 2, the compensation The system 18 uses a plurality of filters 24, no: it is a separate or combined data wave device. # The various mechanical, loudspeaker, and electromagnetic components of the loudspeaker are used to simulate the frequency, interval, and phase response. Can display independent compensation response. 'We combine each filter 24, = per-completion: the response is one or more adjustable parameters 22 = bit signal processor or analog circuit design. Because the speaker behavior is turned on and off It is easy to analyze using digital stem δ generation technology, so digital signal processing is good. Analog circuits have a large attenuation in series connection. So, one: compensation system is compared to f. Only the designer notices that the behavior response is reduced. It is possible to use parallel circuit components only when there is no mutual influence. 479437 V. Description of the invention (15) These parameters 22 will be adjusted to affect one or more parts of the compensation system. A digital system provides Lee The ability to use an editor or calculation relationship that a CD player may set. When the compensation system is controlled by a computer, setting adjustments becomes easy and intuitive. Each calculated and adjusted compensation response can use a parameter variable, characteristic , Design type, size, moving nature, or audio radiation behavior. Time delay, maximum swing limit, knot and standing wave frequencies, boundary reflections, and other speaker properties can be adjusted. The compensation system is also Use. This compensation system uses the least phase equivalence and a string of non-reciprocal: filters to build the entire response system, instead of editing or reducing speakers = as shown in Figure 3f is a typical full response of a compact disc player. And The same response is also shown in the upper diagram of Fig. 4. Figs. 3a to ie show a plurality of individual responses related to the loudspeaker components, and the all responses are combined to form a full response curve. Figs. 4a to "Fig. ^ Individual compensation responses, the compensation responses are adjusted by a plurality of adjustable parameters, such as Ls, Hs, Lx, Hx, etc. = 1. The nature of owing is described in detail below. As shown in Figure 4f, the combination of light response and light response results in a total vehicle response. This combination (Figure 3) and conjugate response (No. ㈣) can produce a glide full response -SP〇nse) 〇 ^ 3a ^ ^ 4a ^ ^; 〇〇t ^

And brother one! The dead-end stops are obvious. The points on these curves determine the frequencies of a Hs, Lx, and 1. When the sound coil length and the moving ton U LS are known, Xu Zengyu, at the beginning, the force of the moon has been changed, or the graph similar to Figure 5 can determine the frequency and phase of Lp.

Page 18 479437 V. Description of Invention (16) " 4 can be determined by similar methods. The flutter and dip graphs show the corresponding inflection, cutoff, and frequency. The setting of Lp amplitude is determined by the power and weight compensation. The self-resonance of the driving action can be measured or calculated to obtain w. . The settings of the phases Q0 and || 0 can be obtained by traditional experimental measurement or calculation, and J is used for audio-coupled movement. Figures 3d and 4d show that the resonance of the bass complement is rarely expressed in traditional models, such as%. The% 4 d! :: high-frequency part shows the audio response error from Wi and is removed. Surrounding resonance can be hunted ["or preset 0 value compensation including resonance of the speaker material is performed:-because of the ear; it is obvious in Figure 3e. In Figure 6 it is dry because it is external 丨I or cone related to the problem of jumping noise caused by ears or noise, such as the 7th _ evening interference, and the frequency knowing π, 戸 λα a Figure 7 / burst or mlssa chart, g§ + tick Paleo and delayed discontinuations have formed high-energy _Songbu frequency compensation notches WCC1 and ^ to eliminate 4t movement. The hidden supplementary frequency and energy storage problems of 苐 4e are shown in Figures 3C and 4c. Behavior and repair: auditory effect. It can increase the subjective balance. The downward slope of the as we see, when each span is 8 _ = to: to do a very good total light repair :. 1 =: about 12 Ύ drives the early frequency response. For example, in the α system, there are two W0 and gg-e related to τ Zan Chuyue's heart, and the second vicissitude response at the lower part of the curve is the disappearance of U θ 丨.驱动 Because of driving 479437 V. Invention description (17) A device works at a very large limit. Other correlation coefficients are obvious, And these curves can be well adjusted. The next parameters allow manufacturing, simulation, or conjugate to a speaker's complex frequency, phase angle, and time response. These adjustment parameters 22 and their operation are approximately zero phase response, Extended bass output, and mechanical sound from a loudspeaker (other parameters can also be manipulated in the same way. #: 广: "Rewards can be performed digitally, or using active or passive analogue :: circuit), analogy Resonance, full state circuit filter implementation. A W King parameters

479437 V. Description of the invention (18) Abbreviation description of the project, related to low cross Lx jurisdictional shooting area, compliance, air loudness, audio coupling, (second change of sound at 4V-6 decibels per octave), high cross Hx group's radiative regional relationship, unnatural, and bl factors. (The first change is at +/- 6 decibels / every octave.) Low Ls design limit, sound balance. (X decibel stop slope) Hs design limit, power control, and effective linear response. (Χ decibel stop slope), T low peak wL enhances the second light transition. Hide low-pass or high-frequency limiter compensation (preset oblique stop slope, X dB peak). The high peak WH enhances the second high-frequency loss change. Hide low-pass or high-frequency limiting filter compensation (preset cut-off slope, X dB peak). ^ Motor BL power, back electromagnetic force, sound change factor. Modulation response and parameter value effects. Bass resonance: W volume, compliance, zone, damping, and bl factors (frequency, q-port adjustment, WB bass adjustment model (traditional) (dual tuning filter model input), mechanical resonance w !, w2 to wn, system components including control electronics Plate, cone, surround, and hemisphere rate, Q, amplitude). Set resonances Wcci, WCC2 $ 丨, Wcc „node mode ·· standing wave, clock mode, delay-light interference Q, amplitude). Coupling factors | Α | !, 丨 A | 2 to | AL Resonance amplitude (adjusted resonance + / -DB). Loss parameters Q〇, Q, JQn Resonance bandwidth (preset or adjust resonance decibel). Wheel hammer DB frequency balance output deviation: factor. Subjective balance and preset equalization. Wave interference Td dimension: ambient, Floor, wall, talker, and listener position. Equalized EQ user operation ... Indoor, other occasions. Power Pw Maximum low-frequency input or output. Analogue or threshold switching detector. Switching mode SW, SW2, and other dynamic programs. : Operation or selection. 1012-3382-??. Dtd Page 21 479437 V. Description of the invention (19) Tuning with the parameters of Λ = i8: 32,… and two operations-the compensation sequence to meet the needs. Since it is related to the type of ，, the adjustment range is required ..., the feeling of type i. The resolution of 8 bits is for this: Two ^ The amount of 15k bits of data in the early day guard is sufficient ... The description of the above parameters is as follows:

The low frequency output of Pl is related to the size of the radiation surface. The compensation of 6 dB / octave caused by the relationship between the courts and the family is # 得 士, # έ 士 4 k, 加, and plus conical motion Limit two ϊ:;! Ίΐ !: ϊ The pressure is smaller than the driving signal. If the compliance is no structure, there will be a flat sound bank. Straight < but Wang Xingui measured, and day 4 # · ^: There are air loudness and mechanical in the different systems, and there are two actual intersections in the speed limit flat to converge to 6 decibels / every Responds to the bass loss of 12 dB / octave. Tongdang said that the lowest response of H is close to the speaker H 'and that the magic parameter turns to ^ Λ Λ. The early position of the parameter is the frequency is Hertz (Hz), and the total number of changes in the car has increased from crying out. In the case of Zhe, it is 6 scallops / mother octave increments. This test word ϊΐΐ — Tone response on # # The value of this parameter can be measured from the input and can be measured by the speaker. The value can also be calculated by Yang Yang, such as cones and coil clusters. The speed of the high-frequency high-frequency output and the size of the radiating surface and the frequency of the stop frequency are usually longer than the speaker output.

4/9437

Every turn is small. Trickling of cone material $ bu_ | shoot. Higher-frequency spokes "== :: When the surface increases in frequency, it will produce a sound s-coil moving group. Finally: the inductance will gather the compliance." Build 6; Dibei :: J rate of the rotary motor = 2: Loss per octave. Combining the actual two minus two, Beisui establishes a reduced audio-compensated control parameter in units of frequency = 讯 讯 应 应. This = the low-crossing method described above is measured or calculated. The maximum increment of charm iikiJ ^ -Ls is limited by the actual release of the crying Λ, Xuan, cone movement, voice coil length, cycle, ‘,》 power, and extension. These requirements will contradict: large phase: large extension and large suspension and suspension components. ; For example, the long sound coil provides a large linear swing and possible rounds:;: 2 rounds with 5 brothers;: Body: 2 with. Continued increase swing required = =: = = = = a can continue, ^ Q J ^ Cloth becomes speaker must utter increase the mobility group and would reduce the effectiveness. If the power of the amplifier is limited, the shell size must be increased. The detailed design of Lx, Wq (described below) and parameters can effectively or extend the design of the speaker response. In addition, the number of Ls meters can change or track different volume control settings and dynamic programs, so that loud sounds 糸, = to operate near the maximum capacity. The Ls parameter is affected by design and construction parameters. The construction parameters in the middle include radiation area, compliance, and

479437 V. Description of the invention (21) Force parameters of the driving motor. For very small interactions, modal parameters can track their changes. The unit of the Ls parameter is +/- dB. It is possible to determine the setting of the h parameter using a test through testing. (i.e., CHS)-The actual maximum performance of the speaker and the power control limit the increment of the highest frequency. In some applications, the Hs parameter can be set in the program source of the speaker to reduce the perceived attenuation of the attenuation. The setting of internal operation can be changed by Hs command. #System inside it: This feature can mask the attenuation of 掸 when the operation exceeds the maximum linear power capability. The unit of the Hs parameter is +/- dB. It is best to determine the Hs parameter by experiments. The cutoff of the resonant bass is often better. Harmony to do this ’and most speakers have self-resonant bass = strong. -Generally speaking, 'larger speakers have lower resonance frequencies, and the combined value of the two is also related to this. The Wl parameter provides a low-frequency resonance lift below the speaker: Self-Resonance (W.). When-the compensation response is reduced, alas, the speaker system operates as electronically and immediately as it gets larger: II. It will not have the rumbling or boxing sound of a miniaturized speaker. In order to reduce J too much, or increase the power, signals lower than I will decay quickly U and include the tracking WL setting in the pass filter. This action is subjective to the f-compensation, because the frequency "cut-off point will be but when it is not heard by human hearing, its call sound can still be masked. Ϊ U low peak is used to adjust and set the maximum Recording and playback = attenuation. In other applications, the high-pass or sub-low-frequency wave 4 is turned off, and the low-pass wave filter that changes or moves parts

/ V. Description of the invention (22) (LP) does not include subjectively including the peak frequency and the unit is male = r; the response of the weekly low-pass oscillating wave is inevitable. The position is the amplitude of the scallop (UL |). The default QL setting is that its resonance cutoff can compensate for the second cutoff tone or reduce the group delay attenuation. Bandwidth ^ System 2; should ... Extend response and peak resonators and track high pass, ^ ^ and write an attenuation program. ~ Value system. Qualcomm's operation is similar to low-frequency peaks. It can also be followed by an i arrangement to establish a weighted response limit. The high-pass 滹 wave is 5¾ μ ciA η 〇 negative drought ^) and the early position is the amplitude u \ |) in decibels. Pre-Qh setting is inevitable. It is the nature of solid structure and sound coil. In the traditional expression f 丨 V 'is the diameter of 3 coils, the winding length, the gap f lux, the pole Vitang camp, and the equivalent of β B LVgdp generator. The component uses a two-acoustic model to make a motor and <m A / c I and other related electric currents related to mechanical properties: the relationship between the 1 and the electromagnetic force of the machine group can be simply expressed. The relationship between leaky sensations and mechanical clusters is sometimes included. These methods use different methods to measure, calculate, or deduce the product, and use different locations to scatter 4a. This can affect the mechanical parameters of the speaker. Compensation system. Therefore, the model can establish a relationship such that the change order of 4 is 2. For example, the specifications of the dagger can be used for tracking or J tuning: it is similar to a speaker with different magnetic weights, sound π King /, factor 0 parameters |. 1 ^^ 0 ^; ^^ Coil length, or other electromagnetic parameters

10! 2-3382-PF-ptd Page 25 479437 V. Description of the invention (23) Swing length, and resistance / inductance or time constant. l ^ &amp; JL (Wn) or self-resonance-this is Yang Qiao μ 6 lack of vibration. This value is determined by the air loudness, mechanical compliance, states, ..., Bass Common = and other design characteristics of the speaker. Unlike the two existing bass models, the audio is coupled to 'ge m'. It has been measured in the compensation system: within Lx: ^ part ’because the parameters and mechanical behavior are converted into RLC resonances to prevent adjustment in Q. , And coupling energy or amplitude 丨 A 丨. The life of +: Two uses I, for the conjugate response. Because the crying rod or °° 4 'refers to the same level of mechanical resonance, but the amplitude is reversed: the response has the phase. We have to adjust W in Hertz. , The number will be very good. Inductive transducers, or instruments and locations that modify the real range of V ~ many speakers are clamoring for the second bass resonance of the A-connector or to change the frequency, including the port, the vacuum tube, and the second two. , Physical sound package, etc. =, Or passive resonance tortuosity, other low research, and achieved; The efficiency of the game is not to increase the rough response of the social small, and other changes-together to improve the bass. The overlap of the effect and the subwoofer is reversed by this compensation system. 479437 5. Invention description (24) and switches are part of the control operation. They are used to define the program status or the response status of the speaker. Sampling or testing constraints determine the switching state of the previous feed / feedback system. The parameter is then selected by this switch or status input. In this method, the compensation system can be controlled by fuzzy logic; or other methods of triggering or changing parameter values unnecessarily. Example: The parameters of the sound control settings, program level, or bass content can be changed. The operation is similar to the sound, but the magnetic control of the control happens to be a major change. Cheap equipment is expected, and with == control logic, you can achieve appropriate idealized adjustments of softness and loudness, and there will be no connection between them: :::-state speaker system \ : ^. Resonance in performance. Unexpectedly, most of them: two P-knives and actions are usually added to the audio output of this speaker. This statement is accidental, and it's the same hemispherical 'cone', and Zhou and Cao's two include the behavior of the parametric equivalent RLC and the resonance of the chassis without the curvature and cluster behavior of each other. Adjustment. The resonance adjustment unit is Sw of Hertz, fixed correction or W-like, Q! To 叱 (described below), and single 1, 2 to ^, the unit is the unit lAIn (described below). A small, low north is the knife ~ circle of 丨 A 丨 丨, 丨 A 12 to respond to a sound collision. Unlike the traditional use of ancient, Zhen # often seems to be able to move around inside, and in time, the other test the response, as soon as the listener rate decreases, it collapses to multi-frequency resonance. Bei ^ Improved the results of the subsidized equalization system from the perception and perception to the signal. t plus / xiao besides the disadvantage and help

/ V. Description of the invention (25) Set resonance (W ^ The hidden compensation program system uses multiple-π 叩 to show you ΛΓ Seoul grabs multiple vibration waves, and the enhanced weight side energy is placed on the frequency, and Xi = Control of parameter control. Delayed node resonance sound result = elimination or removal. Ding Xiao said that the consumptive and loss parameters (U | i '丨 AU, to and% 1, to' Q parameters of certain frequencies can be estimated Its practical, design, and establishment work, the value of the letter i :: 〇 value a is almost impossible to unintentionally be represented by a response graph or simple; ，, using a fixed value on the M boundary, and the preset selection is simple And reduce the value of the adjustment. It is recognized that the limited dimensions of the speaker housing will create a sound wave with a force of 4 t behind the light-emitting surface. Because of the complexity of the modified response, this question should be omitted. Because of the strong electromagnetic coupling Against the anti-wave interference, the speakers of Amber are built by large motors. These systems have two internal obstacles, no, the form of the layer of absorbing material, the special shape of the cavity, or the wire tube to help remove interference. The 纟 -type speaker uses a simple form factor Disturbance 'will eventually cause a response of increasing frequency; = ί: The deterioration of the rule is like a short-term effect, and the damping of the BL will be reduced by small, sexual, reduced steel, or other economic factors. We use delay-feedback -The conjugate inverse of the loop program should be modified. Although it can be used, it is more economical than digital signal processors, and they can be used in a system of f <good equivalent physical behavior, and the problem is fixed. Similarly, ϋf can reduce reflection interference from the floor, table, wall, etc. _ Most of the current listeners prefer the speaker to respond to the current frequency, and increase slowly and slowly. For the lower production cost Loudspeaker

479437 V. Invention Description (26), its response will rise. The compensation system corrects many problems and changes from reducing magnetic weight, improving efficiency, and extending high frequency response. However, the flat audio balance produced by these approximate corrections will sound either muddy or thinner than the current standard tone. The wheel hammer establishes a slope of the compensation response. The operation of the wheel hammer parameter is similar to one-tone control, and achieves the desired triple, called balance. Lx and Hx are manufactured in the same way as traditional response hierarchy methods or procedures. Delay or all-pass-this counts from the high-frequency and low-frequency signals from different directions to 2 MW and multiple-motion signals to compensate the converter _ on the low-frequency field. This adjustment can reduce the group delay and improve the phase matching to one time. Use electronic or physical measurements to do the expansion ... quasi-transient response. This menu can be used to display and adjust parameters on this screen.敫 啊; 浐 二: The amount of lean materials, design representation, response curve provides response and quiet. Remove the traditional traditional practice of building a loudspeaker on the second and second speakers. The experimental machine has multiple mechanical sounds / go: complex Good results. Can use the general digital two A shows an exemplary digital processing. &amp; ϋ Record to achieve completely similar results. This compensation system can also be used. This compensation system can also be used. _ Fc 卜蕾? Underproduction. However, 'unless they are mixed, the :::: circuit damping, or coupling function is not possible, and the common neodymium target tracking adjustment range. Therefore, the Charm circuits do not show their function and analogy. These circuits have parameter adjustment control, but the design system changes the accuracy of component analog parameters. = =

479437 V. Description of the invention (27) and can make the traditional circuit with the synergistic adjustment ability as we described. Gold JUIA ^ -Sallen and &amp; should be high-pass and low-pass filters. Y 町 町 ^ 』U to adjust the peak response in a different amplifier. The IF response is flat, while the progressive gain is limited by + and -18 dB per octave. He Zhancheng ^ Asa loading slope is shown in the figure. However, 'Frequency must be shown in this circuit: For example, because of the wide-band division, f is two but eight, D /; for this example, etc. The surname is as shown in Figure "7, Ya in ^' RC The time constant will change to produce a response curve. The feedback circuit is a precise parameter tone control increase. For the good independence between the electric power, its tuning; Η diagram and Hi-ΥΛ system. Simple analog circuit. The notch depth is maintained constant and Ding I increases with the tuning of the dry ring. For the actual adjustment range, Q is set with the frequency. 卩 is an approximate value. The choice is to provide a reasonable compensation. The chirp wave component is composed of one or more steep hills. The unilateral or bilateral attenuation frequencies can be accumulated. When we properly select its attenuation I amount, increase At the time of energy and bandwidth, the filtering system can eliminate harmful energy caused by slight compromise or change due to sound balance. 4. In addition to the factors of the heart, the attenuation trap has high attenuation and high Q value. Side wave J The compensation or resonance is best determined by the increased response of the sidebands of the attenuation point. For listeners combining rolling response and interference energy problems, single sideband compensation may be more effective. The symmetrical

1012-3382-PF-ptd 1 479437 V. Description of the invention (28) The addition helped to flatten the frequency response. The other types are the same. The device can be: small, with a low Q value, and whether the program is on or off, yes; &gt; the sudden noise of a positive point in the half to eighteen times is a constant. Jie said that Mao Sheng's attenuation can be shifted to the sideband frequency of low Q value by perception. Therefore, for better transient response settings, a pair of tuning adjustments is better than, and a single-side tuning tuning method is better. In order to be able to tune quietly to a useful rate, the crew should carefully tune and adjust the compensation filter. For the nose, if no weight is used, the sidebands of any correction will have a small loss: ί Ϊ 能 ”, but when many corrections are made, 'these losses will be broken ^ α ,, if the response is This phenomenon is particularly noticeable when flat. The weighting method can eliminate the compromise method and make multiple modifiers more effective and reduce the sound 3 ΐ i 'and can also be used when non-internal correction is applied to the speaker; In practice, the weight response notch can be made from the resonance and ringing of the signal. As can be seen from Figure 3, the correction element has a single high-Q notch L, which is centered on the behavior mode, and becomes The boost response of :: Q value is placed at the same frequency to provide complement = core: energy. Figure 13 shows the actual combination of active electric: wave and passive LRC boost resonators on the feedback path. The two circuits establish a 1 ^, heavy-response notch. However, the tuning range is limited, and because it can reduce the transient sound to a low-Q resonance, the low-Q &amp; lift is audible A better choice for W values is to use two low Q value boost responses, Every

1012-3382-PF-ptd Page 31 479437 V. Description of Invention (29) The response can be adjusted and placed on one or both sides of the high-q notch. The 14th unit is shown by four tunable double quadruple resonance or state variable filters. The two tunes of this type of wavelet create high-Q notches in a closed space, and the other two = = 1 provide low-Q energy that is incrementally located at high frequencies and low sidebands. In this state, digital filters can be built with simple digital signal processors and maintain a fixed Q and notch depth over a wide, spectral range. Figure 15 shows the response established by the $ ratio system. The figure has eight coherent I transformers on the same axis to adjust the Wcn or weighted notch parameters. Figure 13 also shows: Scoop circuit. Figure 15 shows two different Q-value settings and the correlation curve corresponding to Figure 1. Aligning the dual-tuned boosts results in a faster and stable transient response. The double-tuning notch shown in the figure provides a similar S. It provides that dead-band or band-reject can go to the manufacturing tolerance produced by another station. We are still stubborn on the side-wave frequencies and it is best to use dual-tuned resonance. When the quality of the speaker is adjusted, the model of the adjustment behavior has similar characteristics. Changes and simplifications are possible. In addition to the compensation-enhancement-like, the two include the two Q-reduction and Q-amplitude settings of the notch-reducing wave. On the combination, there are three traps, two of which have two asymmetric lifting ratios: two on the Q value, one trap, and meanwhile, it can make ^ random noise the average height of the compensation area. θ should be the same or slightly different from the response without correction

V. Description of the invention (30) Delay interference or all-pass-a mixed analogy, a small, convenient adjustable tone delay. Although month:; guys? : Link t is installed like Figure 16 to provide interference-like behavior. ^ This: Road Energy stands from the reflections around walls, tables, and speakers and the approximate conjugate response of the transmission path or system. This voltage can stand up to an inverse comb filter or additional such as time, phase, and positive to remove interference. : Electrical interference. The response of the reflected energy around the enhanced trimmer is-similar to: a good time delay for the speaker to establish; interference shaping; ... :::: number: say processing t; t signal processors can be assembled to be related to physics The number of reflection model two 糸 ϊ ——like, the physical behavior is adjusted by parameter control. The elongation is: The path filtering state is related to the dimensions, absorption surface, and required interference correction. Compared with Figure 16 of the Second County Examination, it can be seen that Td is related to the direct path from the speaker to the listener: the difference between the path and the path. Td # is related to the relative phase return path of the speaker and surroundings. The wall located behind the speaker can also be characterized by ° RC is a large loss with the surface of the coarse sugar or the high-frequency absorption response of J f or comb filtering at the upper frequency. Reduction and correction. R1 controls the adjustment of response or correction. CW =: Xiang + can Λ less, but when ㈣ loses with _ $ 1, the dagger produces the largest additive response. The real amplitude of this circuit generator is dry;! should. The matching or simulation will lose the speaker ', large ° ρ / knife irregular response of small speakers can be caused by

Page 33 479437 V. Description of the invention (31) __ Is it compensated by experimental changes? The setting of the delay usually requires matching the type of f ^ operation. When this is set, the adjustment counts the return waves of the ears. When it is ordered. θ is specialized in the traditional response hierarchy of groups. Figure 17 shows an all-pass or phase-shift network. The frequency is, however, its output is in phase and high frequency, with ^ 疋 flat phase. This circuit does not need to change the frequency output. When it is switched off at low frequencies and switched to CW, the variable control will increase the transient frequency. This telegram: Dong =: II Delays and other related transient responses are useful. , &Gt; Gram test: Two :: Although I still do n’t see or hear with frequency scanner and microphone test, the mechanical confusion of the speaker is often 2. Traditionally, the compensation method is to use a deep notch filter, but ‘neither causes nasal sounds or causes undesired balance variations. This relay wave trap can solve this problem and generate some other problems. The driving state (bass amplifier and tweeter) is driven by a capacitor or a circuit, and the father is more and more up and down, one or two of which The driver can do interference compensation without conscious attenuation. The same application can be made for different positions. When the bit t responds badly, it can be compensated without changing the position. Hidden corrections are made by weighted energy. Because good listening settings do not need to be changed due to correction, a wide range of listening positions will have good sound. This characteristic is particularly useful for sharp-angled speakers in the scene, and only a few corrections need to be made. With preset ideal values for the speaker or construction characteristics, participation adjustment is simple and intuitive. Control Some important properties of the speaker relative to the perceivable effects of sound cannot be 479437 V. Description of the invention (32) Modeling or traditional method of resonance coupling, and it is a non-linear body rigidity and must be considered from the driving sound green. The release of the circle in these behaviors and the phase error of the stereo sound cannot be found or it is difficult to observe by traditional methods. For a specific physical action, the machinery that compensates a large range of speakers is designed to act as a fully active conical element at low frequencies. The active radiation is its primary node, and its nodes collapse. The corrector of this design change can use the complex pass wave receiver to remove the light at the south frequency in time. It will then be reduced, thus maintaining the speaker's ability to remove some distortion of the group delay. The physical dimension of the speed property can be changed. However, the delay effect and the decoupling frequency of some components of the loudspeaker change large parameter values. . Such acts include interference and interference. In addition, other spatial factors related to high-frequency decoupling of moving objects will also cause sound source movement and delay energy differences. Such a destructive response is often a dialect of a microphone-frequency analyzer, an example of a parametric compensation method. Decoupling effects. Such speakers are usually. At high frequencies, this driver is the only one. The cone is intentionally decoupled to attenuate the position of the equivalent radiation source. A linear low-frequency band-limited delay, and when the frequency increases, the phase of the potential delay response order matches. This method is accompanied by the attenuation of the speaker and the cone sound, and is modified by the filter program. The parameter value can be determined experimentally. If it is small or decaying! , You must infer another one of the new parametric compensation methods. Early s. Bühler modulation, this method should use the traditional method: said: the use of speakers, and multi-channel sound system needs to be small: to: Do compensation. The combination of the computer's fast, high-shift cone movements can be heard loudly. In this way, σ 4 increases in our sound reproduction

479437 V. Description of the invention (33) or reduced speed Irradiate the area and worry, and it will fight. When radiation is logical and accurate (Figure 5). Revision late. This increase or even reduction of non-parameters can be difficult to plan. Delay modulation modeled The bass-driven horn processor controls the oscillator at the time and voltage. The analog delay is real. Components and other interface programs are generated by the nodal wave program. Weight compensation is required. It can be true, and it can be used by any human when the voice is replaced. The speaker circle tells us all about the small speaker area, frequency cone movement and speaker response. Non-linear range Linear distortion. The correction. Because it is better to provide a conical movement, it is possible to perform the timing operation immediately. A compensation receiving variable timing distortion wave interference behavioral response can be parametric to the most critical frequency band dynamic line plus bass output speaker system. Cone's low-frequency tone modulation effector will produce a sequence when it and the output Doppler are radiated, which will make a good speaker and an open circuit correct. We are correcting by the model that the complementary gold path is generated from the nonlinear and non-linear movements and decays to its good dynamic hearing. We can use effective high-frequency response to measure ability and response. This effect is quite so that when the sound power of a person playing a loud sound is known, the combined emission parameters can be calculated (the equivalent radiation source with a relatively changed time delay is more stable, the device behavior model and a single dimension Feedforward non-linear correction is an almost-response correction. Extending the time to explain this example is because of making predictions and becoming more practical from digital signals. We have delayed modulation by the bucket brigade of the CMOS type For mobile situations, the use of weight reduction is needed for a good compromise! Position. If program matching, speaker loss, and parameter extension or concealment should be extended. This technique is suitable for

479437 V. Description of the invention (34) ----- ^ However, if the gasket of the speaker is designed to require dynamic compensation, the performance characteristics may not be added. The adjustment table or expected bitmap for this particular system has the same parameters and adjustments as a general-purpose system, that is, the controller has the same characteristics. Therefore, for general and specially designed speaker compensation systems, there are electronic systems for the compensation system to receive commands to compensate for changing speaker models. If required, the system can improve clarity, imaging, and bass extension without changing the sound balance, ie the key signature of the tone. ^ The effect of parameters due to adjustments has been explained by the symbols or abbreviations in the table previously described. In this method, mathematical states, features of circuits, speaker design, and other specific languages are uniformly represented, and the programmer or user can intuitively complete the design. Then, like other response curves, behavioral characteristics, and wavelength relationships that are important for human I and λ, it becomes meaningful when designing or compensating a speaker system. Because changes in design and performance, and the basic relationship, are meaningful and intuitive, compensation behavior is simple. The setting and adjustment of the pitch compensation parameters include the frequency in Hz, the digital Q value, and the rejection in decibels. Because multi-order nodal behavior is possible, the Q value and the repulsion adjustment are similar; ^ a compensation frequency with multiple similar values. The controller can link them so that two or more frequencies with Η z can specify the modification of the nodal behavior. The Be 1 1 mode compensation of the speaker cone provides a good example. Several methods can determine these frequencies. Multiple interference-related nodal waves or waveforms can be observed with sine wave excitation and optical interferometers. Simulation from MLSSA form

1012-3382-PF-ptd Page 3 479437 V. Description of the invention (35) The lacquered cloth picture, as shown in Fig. 7, or use ^^ to show the collapsed set energy. These behaviors include the sequence of the probes, which can be displayed manually. You can do speed sweep manually or at frequency and hide the audio changes during the compensation process. Because of internal reflections, the method of random noise retesting needs to be careful and accumulated. Now, speech = mechanical sound, this is really good. And when several frequency traps with common original J :: $ surgery correction values and common boosting parameters are determined for each frequency, mixing * For the special behavior of the speaker, up. See =. k for assembly compensation. Some procedures are: within the frequency range of the σ dagger adjustment, the _ spectrum has a specific value = and: the control is in a narrow range and there is isolation, so that the mutual influence between each other is: the edge and other special operations have numerical values but do not affect other parameter. In addition, the parameters can be changed at any time with useful adjustment resolution. The selection data of the degree makes it the same as the previous speaker example. The adjustment components are related to the audio properties of the speaker. The control and physics of the system can be improved in the same way. Yes; his similar speakers are also simple and intuitive language. —Small-scale; it can be greatly changed with the system or special information and a few internal values. 2 devices use 16 adjustments to match the complex response, a traditional multiplexing frequency band: [. To enable the system to exceed 100 frequency bands and more control information. — The eyebrow equalizer may require the compensation system to use the slope and predictive corrections to the nature of the speaker.誃 ^ his mathematical function establishment system ′ and replace the file i formula by program commands; 瞀: parameters or characteristics control graphics. Because each program has specific: quality: should, draw negative frequency, amplitude, etc. linearly 479437 V. Description of the invention (36) __ etc. limits the parameter range to a small part of a control space. One three ^ one code can establish an accurate resonance position or a complex weighted notch. $ For an 8-bit code example, the information sequence provides a 128-point logarithmic boat state analysis with an amplitude of 1/2 dB steps. There are +/- 20 dB, and 16 Q settings like * command and control format. For another example, the 84 points or frequency at the sound frequency can be obtained through a notch filter that is set to be severely discarded, and is compensated. The bit map and program command controlled by parameters may only need = degrees, which is more economical. This concise description; Moon Dagger, the installation of the bar code in the compensation system to modify the speaker settings, both of which are used in the program hidden in the data stream to dynamically control = = when the range is reduced, such as hearing aids and Communication system :: With the application of 俨 H ί Γ ί, you can choose effective programs that operate at 384 points or frequency of the audio band, such as interval, notch, fork 1 2 ί delay, etc. The program size is fast with the resolution. The notch can be very sharp and requires accurate tuning. The traditional measured speaker response reveals the program. Unless the compensation system is designed to identify and fix even a good correction will require a lot;

In the case of Wi, and setting problems, the program range of the “# system” was put in second order, and the power of this sequence was //, and it was impractical. When the waver or process is placed in the k-sub-graph, a higher resolution or more is required

V. Explanation of the invention (37) When amending the behavior, the cost of effort or the effort to compensate for the correction is increased. Even if these materials are hidden. When the power fails, it is still beneficial to change the compensation system. Let's take the example of rH. The compensation system has 8 notches,-frequency center; whole ... low frequency increasing crossover and increasing separation. High! 4 echo frequency% acoustic crossover point. Clearly, check out ㈤ The difficulty of synthesis is obvious. The compensation ^ positive values (Lx, LS, Lp, W0, Q0 wide Q-value) establish the conjugate curve. 1 jl 丨 7 Wcci and — Gu Er, similar to compensation The strategy of signal movement in the frequency band can be used for = 忐 xi = frequency. For example, the reduction of the noise caused by the super-cone movement of the bass energy can reduce the noise by reducing the signal of the turn-in speaker. This method can limit the low frequency This is done in response to an actual value. The speaker ^ can also be suppressed, and the weak regenerative bass of A below the stop point makes the program not; the low sound # symbol. The program with a wide dynamic range can force the speaker here The frequency produces a well-known output. Unfortunately, this causes Weak output is often obscured by his voice. Therefore, the extended response of the speaker is useless most of the time, it only increases the attenuation and consumes the power of the amplifier. It is said that the weight response of the compensation system is suitable for low frequency and high Bandwidth f or the cut-off point of the oscilloscope. If performed carefully, the sensory two should not be changed and the sound can be stretched before filtering. Even if the same method is used, large systems are expected to require different Methods to regenerate high power, audio, or subsonic energy. The compensation system combines a speaker model and a conjugate program to establish a

1012-3382-PF.ptd 479437 V. Description of the invention (38) Low-group-delay attenuation audio is private, and the peak response not only complies with the two-phase zero-wave generator. For such application delay 'and it becomes the allowable limit ::: conventional' and it is also adjusted to help the group to choose the hidden compensation pad of the neutral band switch. Diameter and minimum sharpness. The compensation system can change its cut-off tone and control the m-rate grade feature to allow the compensation of n to be changed by sound. In this method, the change of the condition is tracked or adjusted. The front wheel A% H ~%, 1 ^ See that can force more power to attenuate the sound ... Because the sound energy is quantitative, a small amplifier can produce Same recording and playback volume. Li 97 complements important business considerations. The loudspeaker can be designed as an advantage of this system. They use shorter sound coils and brighter weights: solid cones, ’to produce higher amplifier sound output. Performance is a comparison. Does the sound output increase with frequency? ;: =, As well as the Sibei Qianv door and right designed in this way, the sound of the loudspeaker will be proud. However, these changes are caused by corrections made by the compensation system, the loudspeaker can play louder, and a lower power amplifier can be used. · If you choose a substantial bass extension, the peak or patch moves with this strategy. Program settings and band edges: ~ ', said frequency will change with the bass content of the program data, the inverse value of the amplifier, and the ability of the speaker to move. When the high recording energy is strong, the peak response will be improved to maintain one person's w μ gate and attenuation performance. , Quasi-hold ... cone movement Most entertainment environments have equalizers, tone controllers, and other

479437

Or the initial best sound system is to equalize the internal objects. This ensures that the sound is removed and the low frequencies are not used by other systems. Because the difference is that the speaker is not connected by the speaker. Because it is harsh, the speaker 0 operates in response to the speaker in series or the appropriate action can also be performed. These sounds are mobile energy-enhancing features. The application of compensation will not pick out the inappropriate ones, which will disperse, cognitive systems, etc., and generate the audio, and the system will ask or guess. The equalization of attention, and the method of bitization. Rumble or visibility can be changed to provide an ideal measurement of the microphone compensation system. The sense of mechanical position that can be damaged can enhance the construction of them. The measurement of the light shooting wind can have more phase response sound. Knowledge is a model of good frequency. The effect of compensation is the response to the quantity response effect. String human two good. No buzzing sounds For this compensation system, specific analog function block circuits can make practical systems. The response range of the adjustment parameters can be estimated by the conjugate model ^, so parallel signal paths and multiplexing sharing functions are possible. A signal path requires relatively few active components, reducing cost and distortion. It is possible for the all-pass filter to produce some time-dependent corrections, but other delays are useless. The analog and digital signal processing implementations of the demonstration speaker compensation system are shown in Appendix A and Appendix B, respectively. Other applications of compensation are described below ^. A multi-system communication system encounters unwanted narrow-band noise or tones when cutting or changing frequencies. Some waiting time is often necessary to enable a subsystem to define and track the tone. When it is not practical or possible to build a phase-free signal to remove disturbances, the compensation system can achieve

1012-3382-PF.ptd Page 42 Effective effects. Generally speaking, the sharp tracking disturbance is possible: if there is] = the weight parameter of the mobile parameter responds to the notch number, then the correction will be effective when the smart communication system that reduces or changes the different resistance frequencies has a microphone,: of. Generally speaking, the device is used as an audio reproduction system, and the conversion of the machine and the speaker corrects the annoying response of a system. One = equal component has any position that can be individually compensated or. They can be controlled = two processors can be placed in the signal chain g, the signal path, and the human hearing two = two to compensate for many hybrid components such as transforming this system and controlling parameters. To create or access. Like other methods, the hidden code in the signal path is used for authentication. Because the effective address of the compensation system can be used to transmit the response of these resources, this correction procedure must provide some information to establish a very complex telephone system for the hearing impaired, many systems. For example, computer speakers and processors can be connected externally or not. For these systems, the processing is programmed as part of a 4m device. Therefore the compensation system is connected. Some, all are not =, and can be movable outside. Many two need to be placed in the device. of. People often have difficulty with β-force because of the damage caused by auditory nerves. : T sound ’: It ’s hard to get a nerve response in a crowded noisy room. 4 It ’s extremely irritating to the big and strong stimulus. Hanging on can cause uncoordinated and annoying sound. Xigu The feedback path to the brain collapses, so that the movement that can be started can be measured by the medicine February: ",: 2 The mechanical type caused by the stimulus, the formula finds its sacrifice parameter weight trap filter

Page 43 479437 V. Description of the invention (41)] Yes, the adjustment to the same frequency to prevent the sound transmission wheel from increasing the weight of these fine right sets is set individually, and the loss will be reduced. In a typical case, the notch: = becomes a resistive characteristic band for wider sound pressure. Compensated edge = changed to reduce the intelligibility loss of other sounds. Design: ,, first: perform this correction more efficiently, and its specific uses less power than the general-purpose response profile method. Recording can be done by removing the known frequency from the notch. Loss = this same recording needs timing, then m looks for this known = rate. We can expect irregular beats, noise, and unpredictable statics, and many problems arise. Recording may use the operating frequency and time of the decoded safety notch. The degree of chronological dependence will be reduced, and several levels will be changed. Initially recorded parameters during recording and playback. Just like the communication system described earlier-the pro-wave device will be quiet or only possibly hidden, it is similar to replenishing 俨: =: another frequency. For this application, parametric control provides an effective safety system. … The key frequency ¥ beats and hides the reinforcement of big sounds ~ installation, headphones, and; will have both feedback and feedback. -The microphone passes through Bing Jiu, and it is added to the original signal to generate shock. The complex digital signal processor of the radar can pick out the internal phase relationship to clear the transmission of the sound and eliminate the feedback. Frequency shift:

479437 V. Description of the invention (42), the feedback phase of the machine is used to prevent slowly increasing laughter or high-Q coupling loss ringing. Sometimes the access time of these methods interferes with performance or synchronization. The tunable weighted notch can be set to a high sensitivity to the frequency to eliminate oscillations, but it does not cause a few easy to each other. The electronic system of the CD-ROM, the description of the pieces, the combination of mathematics, and the coefficient values have determined the mathematical system. This parameter indicates the number of other devices, which can be easily downloaded, downloaded, or installed. Although the present invention limits the present invention, within the scope of God and the scope, it will be accompanied by a significant loss of sound. The effects of dual-tuned or multi-tuned circuits ’and the built-in static frequency band will make feedback cancellation more tolerable. Meta-information may be made and their design is electronically or specifically modified on a turntable player or implemented in between to control, electronically and not be used in the spirit of the present invention's scope of protection-one that works in the digital or analogue range (or may incorporate internal users Adjustment and compensation model, measurement data, and human favorite and merged design. When the parameters are hidden, it is determined that a group of recognized values can program the sum of speakers of different depths and sizes to be modified independently. The model can be downloaded. To an optical signal processor. This operation can be upgraded at any time. It can be activated by a computer barcode and remotely mounted on a dedicated player / speaker ^ It has been disclosed above in a preferred embodiment, but anyone skilled in this art You can make changes and retouching without disassociation. Therefore, the scope of this patent shall prevail.

Claims (1)

The scope of the application for the patent application is to change = input to an audio reproduction device-electronic voice signal generation = = to combine the individual responses of Γ to define the audio and then to temporarily model the friction: Yes, the response of the mother-in-kind includes frequency, Time, phase, or at least one of the following, the device includes: Change filter: ΐ t change the filter '▼ to modify the individual response of the device, these repairs and i supply; to receive the electronic voice signal' change the electronic The speech signal, U, the electronic speech signal of the reproduction device; and the U 'adjustment parameters' are each related to at least-the modified filters, which are the responses of the entire other filters; should, ^ a h The iE ′ generates a plurality of individual conjugates. 2 The response phase of the conjugation is related to at least one other response device. The device described in item 1 of the patent application filed by the Department of Health, wherein the audio remagnetism ^ A is related to at least one of the mechanical, auditory, and reading of the audio reproduction device by individual responses. . Among them, the filtering, among the filtering, among these modifications, among others, the openings 3 3 The device described in item 1 of the scope of patent application is defined by digital signal processing. 4 The device described in item 1 of the scope of patent application 15) is defined by an analog circuit. 5 · The device described in item 1 of the scope of the patent application has no interaction with the device (η〇η-interacting), and it should be installed in the factory 6! Please refer to the device described in item 1 of the scope of patent, where the amendments i should. The devices are combined to form a conjugate that is responsive to the full response of the audio reproduction device. 479437 6. Scope of patent application The modified filter includes a cut-off filter (cut-off fUter), and the coefficients that adjust the frequency response of the cut-off filter include the peak frequency (frequency), amplitude, and Q parameters. 8. The device according to item 7 of the patent application range, wherein the frequency response of the peak frequency, amplitude, and Q parameter modification cut-off filter is at least in a low and high frequency range. "&Quot; ^ 9 · The device as described in the first item of the patent application, wherein at least one modification filter includes a constant slope equalizer (c0nstant ^ lQpe &quot; equalizer ·), and the constant slope is adjusted, etc. The parameters of the frequency response of the chemist include the crossover frequency and the boast shelf parameters. I 0. Pei Zhi as described in item 9 of the patent application range, wherein the crossover frequency and the interval parameter change the constant slope equalizer's frequency response in at least a low and high frequency range. II. The device according to item 1 of the scope of patent application, wherein at least one modification filter includes a parametric notch filter (parametric n () tch f 11 ter), and the parameter of the parameter notch filter frequency response is adjusted Includes notch frequency, amplitude, and q parameters. 1 2 · The decoration according to item 1 of the scope of patent application, wherein at least one modification filter includes a parametric notch-boost filter, and the parameter notch-boost filter is adjusted Frequency response parameters include notch frequency, amplitude, and Q parameters. 1 3 — An audio compensation system used to change an input to an audio reproduction with related behavioral characteristics On page 47, the electronic voice signal set in the scope of patent application, the system includes a reproduction device model, which has a plurality of analogues of at least one of the audio signals, and the i-wave characteristics of i should be widely used. The full response of the 疋 a model, each response includes at least one of frequency, phase, or transient response; and Cheng Cheng: modify the response of each filter, and convert the filter A conjugate filter that is conjugated to one of the original responses of the filter is a system described in item 13 of the patent application, wherein these lines are determined. The individual components of the Japanese Λ regeneration device are bounded by the system described in Item 13 of the declared patent scope, in which the indiv'i (groups of Ual components of the audio reproduction device) The system described in item 13 of the scope of patent application for the wave device, wherein the filters are defined by digital signal processing, and the controller includes an electronic wave counter. The system, which includes these components. 'The system is defined by the circuit, and the controller includes an adjustable circuit, the system described in item 13 of the patent application scope, the neutral K speaker, ...: . ^ And the calculation of 忒 parameter values is based on the actual speaker Page 48 479437 VI. Application scope of patent 19 • The system described in item 18 of the scope of patent application 'where the actual characteristics of the speaker include cone, coil mass, air loudness (air volume), mechanical compliance (fflechanical compl iance), radiation area, damping (damPing), moving mass (moving mass), and motor characteristics are less. 20. The system according to item 3 of the patent application scope, wherein the audio reproduction device includes a speaker, and at least the filter includes at least one related adjustable parameter, and the parameter value is obtained from a standard speaker model. . 2 1 · The system according to item 3 of the scope of patent application, wherein at least one filter includes at least one related adjustable parameter ', and the parameter value is determined experimentally by a standard test measurement. 22. The system according to item 3 of the scope of patent application, wherein the controller is equipped so as to control the setting of at least one other parameter when setting a parameter adjustment. 2 3. The system according to item 22 of the scope of patent application, wherein the audio reproduction device includes a speaker and a parameter that controls at least one other magnetic structure parameter and a voice coil related to the speaker. 24. The charging device as described in item 3 of the patent application scope, wherein the controller monitors program conditions of the audio reproduction device and sets at least one parameter value based on the program conditions. 2 5 · The system described in item 24 of the scope of patent application, wherein the program conditions include a volume control setting, program level (program 1012-3382-PF * ptd Page 49 479437 6. At least one of the scope of patent application) and bass content. 2 6 · The system described in item 13 of the scope of patent application, wherein one of the wave filters includes a weighted compensation notch filter 〇27 · as in the scope of patent application 26 The system of clause 1, wherein the filter comprises a single-tuned weight-compensated notch. 28. The system of claim 26, wherein the filter family includes a dual-tuned weight-compensated notch. 2 9 ·· An audio system, including: an audio reproduction device with mechanical, auditory, and electromagnetic behavioral characteristics; a signal source (source), outputting an electronic voice signal to the audio reproduction device model, the model has a complex number A complex filter that simulates at least one of the mechanical, auditory, and electromagnetic behavior characteristics of the audio reproduction device, each wave filter has a related response including at least one of a frequency, time, phase, or transient response, the The model outputs the electronic voice signal to the audio reproduction device; and the controller 'modifies the response of the filter so that the model is converted into a car-sharing model with a plurality of filters, and the wave filter has a car-sharing in the filter Of the original response. 3, The system according to item 29 of the patent application scope, wherein the filters are defined by digital signal processing. 31. The system as described in item 29 of the patent application scope, wherein the wavelets are defined by analog circuits. 6. Application for Patent Application 圏 3 2. The system is based on the system described in item 29 of several patents, where the mixed skin states do not interact with each other. The τ 忒 post-finding filter includes the system described in item 29 of the patent application, where at least-the modification includes the second wave: 'wave filter' and the frequency response of the cutoff filter 34 対-frequency, amplitude, And modification of Q parameters. The filter includes the system described in item 29 of the second range of patent application, in which at least-the rate response modification spoon ^: equalizer, and the constant slope equalizer frequency 3 5 Modification of crossover frequency and interval parameters. The wavelet includes the modification of the ^ rate response as described in item 29 of the application section I, including the pair m 'and the notch filter frequency of the parameter, and modify. Qing, * The system as described in item 29 of the scope of patent application, at least one of which is Γ; === lift wave filter, and the parameter notch lift filter 'wave = drought drought% should be modified to include the notch frequency , Amplitude, and Q parameters are modified. v 37. A method for modifying an electronic voice message input to an audio reproduction device, characterized in that a plurality of individual responses are combined to define the one tone: the full response of the regenerative device, and each individual response includes frequency and time The method includes the following steps: using a complex filter to simulate the individual individual responses of the complex; adjusting the responses of the complex filters such that for each wavelet, The unitary response includes a response conjugated to the one; τ I inputs the electronic speech signal to the filters Page 51 479437 38. The method as described in item 37 of the scope of patent application, wherein the plural individual responses of the tone a Dansheng device are at least related to the mechanical, auditory, and electromagnetic behavior of the tone A reproduction device. 39. The method according to item 3 of the scope of patent application, wherein the plurality of filters do not interact with each other. 40. The method as described in item 37 of the scope of patent application, wherein the plurality of adjusted response responses are combined to form a full response conjugated to the full response of the audio reproduction device. 41. The method according to item 37 of the scope of patent application, wherein at least one filter includes a cut-off filter and the step of adjusting the frequency response of the chopping wave filter includes setting at least one of a peak frequency, amplitude, and Q parameter step. 42. The method according to item 37 of the scope of patent application, wherein at least one filter includes a constant slope equalizer, and the frequency response step of adjusting the constant slope equalization step includes setting at least the crossover frequency and the separation parameter. - A step of. 43. The method according to item 37 of the scope of patent application, wherein at least one filter includes a parameter notch filter, and the frequency response step of adjusting the parameter 滹 includes setting-notch frequency,-amplitude; Benefit at least one of the steps. Zhizhong 44. The method described in item 37 of the scope of patent application, a small to medium wave trap is used to raise the wave filter, and the parameter is adjusted to: lift; the frequency response step of the wave source includes setting a notch frequency, neck- 479437 6. In the scope of the patent application, 45. A method for changing an electronic voice signal input to a tone reproduction device with related behavioral characteristics, the method includes the following steps: Simulate at least one behavioral trait of the audio reproduction device. 'Each filter has a correlated response including at least one of frequency, time, phase, or transient response; and modify the filter's response for each filter. , So that the filter is converted into a common vehicle filter having a response common to the original response of the filter. Where the behavior Where the behavior 〇 Where the audio 46. The method described in item 45 of the scope of patent application, the characteristics are defined by individual components of the audio reproduction device. 4 7 · The method according to item 45 of the scope of patent application, the characteristics of which are defined by the individual component groups of the audio reproduction device 48 · The method according to item 45 of the scope of patent application, the reproduction device includes a speaker, and The at least one filter includes at least one related adjustable parameter, and the step of modifying the response of the filters includes the following steps: calculating at least one adjustable parameter value based on the physical properties of the speaker; and — setting the parameter to the Calculated. 4 9 · The method according to item 48 of the scope of patent application, wherein the physical properties of the speakers include at least one of the characteristics of a cone, a coil cluster, air volume, mechanical compliance, a transmission area, damping, a moving mass, and a motor. One. 5 0 · The method according to item 45 of the scope of patent application, wherein the audio 1012-3382-PF-ptd Page 53 VI. The scope of the patent application includes a loudspeaker, and at least one filter includes at least one related adjustable parameter, and modifying the response of these filters includes the following steps: Obtaining at least one adjustable parameter from a standard speaker; and setting the parameter to the acquired value. 5 1 · The method according to item 45 of the scope of patent application, wherein at least the complex filtering state has at least one related adjustable parameter, and modifying the wave-like response includes the following steps: · Experimentally using standard test and measurement methods To determine at least one related adjustable parameter; and set the parameter to the determined value. Method, 52. The parameter setting as described in item 48, 50 or 51 of the scope of the patent application further includes the step of adjusting at least one father in response to the setting of another parameter. Recognition method, its% in r 1 β, and cake-like shape / brother, 矣 少 -parameter where the money and low 53. If the patent application scope item 48, 50 or 51 includes the following steps: Monitor the The audio reproduction device is set at least based on the at least one program condition ^ 54. Gross as described in Item 53 of the scope of patent application:%; low volume (baSS-type conditions include at least one of a volume control setting, program level I content) . 1012-3382-PF-ptd 54th tribute