Method and system for adapting the measured EQ response of a sub-band to the pre-calibrated EQ response of the EQ of an audio system of a specific room

Abstract

A method and system for pre-calibrating an equalizer (EQ) response of an audio equalizer EQ to adapt a measurement sub-band of an audio frequency response equalization (EQ) response to a specific room's audio system, utilizing a modified genetic algorithm to adapt an off-the-shelf audio system (such as in a high-end television) to a given specific room or other physical location to present a specific or unique listening environment with a set of acoustic characteristics. The audio system is adapted to the given room by determining an IIR-based EQ solution via an iterative technique, including an iterative technique based on a genetic algorithm applicable to audio frequency response equalization applications. In one variant, the audio system is adapted to the specific room, adjusting the EQ over a microphone bandwidth while maintaining a factory-calibrated EQ response over the remaining bandwidth.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method and system of pre-calibrating an equalizer (EQ) response of an audio equalizer EQ of an audio system to a specific room's audio frequency response equalization (EQ) response of a measurement sub-band. BACKGROUND

[0002] Enhancing factory tuning of a television (TV) to address the acoustics and conditions of any specific room presents several challenges to the designer. The processing bandwidth of the TV's system on chip (SoC) must be considered. Factory-based tuning typically uses a high-speed laptop, while home tuning is typically limited by the bandwidth available to the TV SoC. Furthermore, factory tuning is typically based on recordings made in the audio near-field in a controlled environment. These recordings are used to compensate for speaker frequency response limitations. However, home tuning typically involves recordings made at the listener's location or through a microphone mounted on the TV, in a room with furniture and doorways, uneven surfaces, and typically windows, i.e., a relatively "hostile" audio environment. These recordings can be used to compensate for both speaker limitations at the recording location and room effects. For home tuning, the recording device sample rate typically varies between 8 kHz and 48 kHz. Also, the recording bandwidth, e.g., in a TV remote control, is typically limited to the speech band (300 to 3 kHz).

[0003] Finite impulse response (FIR) filters are commonly used in audio frequency response equalization (EQ) filter applications. Once the impulse response is known, the FIR solution for compensating for the frequency response of a random listener position in a room is well defined mathematically. The resulting FIR filter is typically high order, requiring a large amount of processor bandwidth to implement. Infinite impulse response (IIR) filters, which have equivalent (equivalent to FIR filters) performance, are much lower order and thus require less bandwidth to implement. Once the IIR filter coefficients (EQ solution) are available, they provide a more efficient method to compensate for non-ideal room responses. IIR filters are recursive, with both zeros and poles, and thus are prone to stability issues. Deriving an IIR-based solution for a random room response is a complex mathematical process that requires careful handling of the filter's zeros and poles. SUMMARY

[0004] Embodiments of the present disclosure provide the ability to efficiently adapt a factory "off-the-shelf" audio system to any given room.

[0005] One aspect of the present disclosure relates to adapting an audio system to a given room by determining an IIR-based EQ solution via an iterative technique, the iterative technique including an iterative technique based on a genetic algorithm applicable to audio frequency response equalization applications.

[0006] Another aspect of the disclosure is directed to a system, method, and algorithm that adjusts EQ across the microphone bandwidth when tuning an audio system to a specific room, while maintaining the factory calibrated EQ response across the remaining bandwidth.

[0007] According to one aspect of the invention, there is provided a method for adapting a measured sub-band of an EQ response to a pre-calibrated EQ response of an audio equalizer EQ of an audio system for a specific room, the method comprising: a. creating an EQ target by splicing an ideal response of the measured sub-band into the pre-calibrated EQ response, thereby forming a spliced response; b. defining a transition area on both sides of the spliced response; and c. matching the level of the spliced response to the pre-calibrated EQ response using a statistical technique.

[0008] According to another aspect of the invention, there is provided a system for adapting a measured sub-band of an EQ response to a pre-calibrated EQ response of an audio equalizer EQ of an audio system for a specific room, the system comprising: a memory comprising computer executable instructions; and a processor coupled to the memory and operative to execute the computer executable instructions, the computer executable instructions causing the processor to perform the steps of the method as described above. BRIEF DESCRIPTION OF DRAWINGS

[0009] The manner and process of making and using the disclosed embodiments can be understood by reference to the drawings, wherein the principles of the concepts described herein are illustrated in diagrammatic form. It should be realized that the components and structures shown in the drawings are not necessarily drawn to scale, emphasis instead being placed upon illustrating the principles of the concepts described herein. Like reference numerals in different drawings denote corresponding parts. Furthermore, embodiments are shown in the drawings in an example and not a limiting manner, wherein:

[0010] Figure 1 is a plot showing the actual versus target speaker / room response for a given listening environment;

[0011] Figure 2 is a plot showing Figure 1 the difference between the target and actual response of the speaker / room of

[0012] Figure 3 shows an example of chromosome encoding for an exemplary embodiment of the disclosure;

[0013] Figure 4 shows an example of a cross technique for an exemplary embodiment of the disclosure;

[0014] Figure 5 is a plot showing an example of identifying three largest error peaks from the best current solution according to an exemplary embodiment of the disclosure;

[0015] Figure 6 is a schematic showing the addition of new PEQ bands to each chromosome;

[0016] Figure 7 is a block diagram of an exemplary method of adapting an IIR filter to a given room according to the present disclosure;

[0017] Figure 8 is a plot showing the EQ response of an initial factory calibration, the ideal recording derived response, and the resulting spliced target response according to exemplary embodiments of the present disclosure;

[0018] Figure 9 is a plot showing a measured speaker / room response with a steep roll-off for an 8 kHz sampling rate in a recording device according to exemplary embodiments of the present disclosure;

[0019] Figure 10 is a plot showing an example of a modified upper transition region for a recording with a steep roll-off according to exemplary embodiments of the present disclosure;

[0020] Figure 11 is a block diagram of an example of another method of adapting an IIR filter to a given room according to the present disclosure; and

[0021] Figure 12 is a schematic diagram of an example computer system that can perform all or at least a portion of the methods, algorithms, and processes according to the present disclosure. DETAILED DESCRIPTION

[0022] The features and advantages described herein are not all-inclusive; many additional features and advantages will be apparent to one of ordinary skill in the art in view of the drawings, specification, and claims. Moreover, it should be noted that the language used in the specification has primarily a purpose of readability and guidance not of limitation. The present subject technology is applicable to many embodiments. The following is an illustrative rather than exhaustive description of the scope of the subject technology.

[0023] To equalize the audio response of a television audio system, a parametric equalizer (EQ) can be used. Each band of a parametric equalizer has an adjustment for gain, center frequency, and quality Q (bandwidth adjustment). When multiple bands are used, the parametric EQ (PEQ) has many degrees of adjustment. It can be a challenging problem to find an EQ setting for a given room that can provide the desired overall system response. The present disclosure provides novel systems, methods, and genetic-based algorithms (including variants), as described in further detail below, that have been shown to be effective and efficient for finding the best solution for equalizing a given room or other enclosed physical space using IIR PEQ filters (e.g., second order in exemplary embodiments). Using the genetic-type algorithms described herein provides increased effectiveness and efficiency for tuning or adapting an audio system to a particular listening environment.

[0024] One aspect of the present disclosure relates to a method / algorithm for adapting the IIR filters of a given audio system to achieve the ideal EQ response of the audio system for a given room or other listening environment in which the system is used, across the entire range of the system output (or "full frequency band") using a genetic algorithm approach. Details of the individual steps for such a method are described below with reference to Figures 1-6 Figure 7 Details of the corresponding higher-level algorithm are described below with reference to

[0025] As an initial step, the impulse response of the system loudspeaker in a particular room is obtained (e.g., measured or recorded) using an excitation such as a maximum length sequence (MLS). Other excitations can additionally or alternatively be used, such as a log-sine sweep, chirp, etc. The impulse response is then converted to a frequency response using an FFT, which can then be (optionally) smoothed and resampled using a log frequency domain spacing. The frequency response of the system is then compared to a desired or target frequency response to derive the ideal EQ response. Figure 1 is a plot 100 showing the actual response of a given listening environment versus a target loudspeaker-to-room ("speaker / room") response.

[0026] The difference between the target response and the speaker response gives the ideal EQ response, as shown in plot 200 in Figure 2 The ideal EQ response can be derived or calculated by subtracting the actual frequency response of the system from the ideal or target frequency response. When applied to a particular system, the ideal EQ response converts the actual frequency response of the system to the desired frequency response, thereby adapting the audio system to the particular room or location. Embodiments of the present disclosure are used to quickly and efficiently find the PEQ parameters that most closely match the ideal EQ response.

[0027] ​To find the PEQ parameters that most closely match the target EQ response, the example algorithm of the present disclosure utilizes a new genetic algorithm(s) that works by creating a population of chromosomes, where each chromosome is a binary number that represents a set of encoded PEQ parameters, where this set of PEQ parameters spans the entire range of the system output. Each chromosome represents one possible EQ solution. For example, in a preferred embodiment, each chromosome uses 10 bits for gain, 8 bits for center frequency, and 8 bits for Q for each frequency band. The total number of bits and the bit allocation for each chromosome can vary depending on the desired accuracy and / or efficiency.

[0028] Figure 3 An example 300 of chromosome encoding for an example embodiment of the present disclosure is shown. This example shows a chromosome containing N frequency bands, each represented by one PEQ filter. It should be understood that, for example, if the frequencies and bandwidths are fixed (as in a graphic EQ approach), then the chromosome in that case can contain only gain information.

[0029] An initial population (first generation) can be created by choosing random values for each chromosome bit. The population size is adjustable. A larger population will tend to have greater diversity and can lead to better solutions. A smaller population will require less computation and can be more efficient. The next generation can be created by the following. The "fitness" of each chromosome is calculated by computing the mean squared error (MSE) of its frequency response relative to the ideal EQ response. Other fitness metrics can also be used in addition to or instead of MSE, such as mean absolute error or maximum error. The number of frequency bands in the chromosome can be adjusted as the genetic algorithm progresses. Using a seven-band EQ as an example, the process can be started with only two frequency bands and additional frequency bands are gradually added (up to a total of seven) in future generations.

[0030] When MSE is used as the fitness metric, chromosomes with lower MSE are considered to be better solutions. In audio applications, certain frequency bands are more critical than others. Therefore, the weights of the critical bands are preferably greater when calculating the fitness of a chromosome. The genetic algorithm will naturally reduce the EQ error in these bands to a lower level and will not use unnecessary iterations, filters, or chromosomes to optimize bands that are less important from a listener's perception perspective.

[0031] In the preferred embodiment, the chromosomes with the best fitness are transferred intact to the next generation, so that the solution never reverts. The fitness score of each chromosome is then used to measure how likely it is to be a parent to the next generation. Higher fitness means greater likelihood of selection. In this example, 30% of the population are parents to the next generation. (Of course, the percentage of the population that becomes parents to the next generation can be adjusted / selected as desired.) Each child is then created by selecting two random parents and crossing their chromosomes. Figure 4 An example 400 of a crossover technique for use in the example embodiments of the present disclosure is shown. As shown, Figure 4 As shown, an even crossover is performed by selecting each child from one of the two parents at random. Other techniques, such as single-point or two-point crossover are also possible. Single-point crossover can be accomplished by selecting a random index and concatenating the value of parent 1 before the index and the value of parent 2 after the index.

[0032] Next, as part of the genetic algorithm, random mutation is applied, for example, by giving each child chromosome bit a small probability of inverting its value. The fitness of each child chromosome is then measured, and the process is repeated until termination.

[0033] In audio applications, it is reasonable to tune a total of seven parametric EQ bands, for example, as used in the preferred embodiments of the present disclosure. However, for the start of the described process used in the example embodiments, only two PEQ bands are initially tuned. This feature of the audio EQ fitting algorithm takes advantage of the fact that a smaller population will require less computation. Then, after a fixed number of generations, an additional band(s) is added until all seven bands are finally included. This is done to provide more reliable results in fewer total generations. While reference is made to seven parametric bands, a different number of such bands can be used in other embodiments or applications. Figure 5 A graph 500 showing example identification of three largest error peaks from the best current solution according to the example embodiments of the present disclosure.

[0034] When a new PEQ band is added, the new chromosome bit is given an initial value calculated as follows. As shown, Figure 5As shown, three peaks with the largest errors (i.e., the largest error peaks) in the best current solution are identified. The PEQ parameters are calculated, which provide a narrow (high Q) corresponding gain that drives the error to zero at these points. This application-specific modification allows each introduced audio EQ band to lower the chromosome fitness even before the adaptation starts. As a result, it enables the adaptation of the PEQ parameters - and thus the IIR filter values for adapting a given specific audio system to a given room or location - to converge to a suitable solution faster compared to the prior art. Adding EQ bands over time in combination with the fitness-based initialization of the added bands provides a method to improve the effectiveness and efficiency of the genetic algorithm in audio EQ applications, which adapts a given audio system to a given room or location.

[0035] One of the three sets of parameters is attached to each sub-bit. When the algorithm recovers, the new PEQ band can evolve freely by itself. This approach is used to speed up the evolution process instead of waiting for a suitable value to be found randomly.

[0036] Figure 6 is a schematic diagram 600 showing the addition of a new PEQ band. Figure 6 The addition of one new PEQ band is shown, where P is the size of the population and M is the number of PEQ bands currently being tuned. M is less than the number of available bands in the implemented EQ.

[0037] When calculating the fitness value of a chromosome, the bits are converted to their frequency, gain, and Q values. They are considered as Gray-coded values when they are converted back from binary values to parameter values. Gray code (Gray coding) is preferably used to make the evolution smoother, faster. In Gray code, the next value in the sequence can always be reached by changing only a single bit (as opposed to binary coding).

[0038] Table 1: Gray code vs. binary

[0039] Decimal Binary Gray 0 0000 0000 1 0001 0001 2 0010 0011 3 0011 0010 4 0100 0110 5 0101 0111 6 0110 0101 7 0111 0100 8 1000 1100 9 1001 1101 10 1010 1111 11 1011 1110 12 1100 1010 13 1101 1011 14 1110 1001 15 1111 1000

[0040] In a 4-bit example, the binary value of 7 is 0111 and the binary value of 8 is 1000. Using Gray code, the value of 7 is 0100 and the value of 8 is 1100. If a gain value needs to evolve slightly from 7 to the nearest value 8, it needs to flip all 4 bits in binary but only one bit in Gray code. Using binary code, the adaptation would tend to get stuck when trying to evolve from 7 to 8. Using Gray code, the closer value is easier to reach by mutation. According to the present disclosure, Gray coding the PEQ parameters provides a means to reach a satisfactory solution using fewer iterations and smaller population size when the genetic algorithm is applied to an audio EQ application.

[0041] Once all the required frequency bands have been added and when the performance metric of the algorithm or termination criteria is reached, e.g., a fixed number of generations has been processed or when the target MSE is reached, the algorithm terminates or is completed (or can be considered to have terminated or completed). The generated PEQ parameters (returned from the algorithm upon termination) are correspondingly optimized or tailored to the particular room from which the system impulse response was originally obtained.

[0042] Figure 7 is a block diagram of an exemplary method / algorithm 700 for adapting an IIR filter of a given audio system using a genetic algorithm approach across the entire range of system output to achieve a desired EQ response of the audio system for a given room or other acoustic environment in which the system is used. The method 700, which can also be considered an algorithm, includes obtaining a system impulse response of a particular audio system across the entire range of system output for a particular room, as shown in step 702. The impulse response can be converted to a frequency response, as shown in step 704. The frequency response can be compared to a target frequency response of the audio system, as shown in step 706. A desired EQ response of the system across the entire range of system output can be obtained based on the comparison of the frequency response to the target frequency response, as shown in step 708.

[0043] Continuing the description of the method 700, a genetic-based adaptation algorithm can be implemented that creates a population of chromosomes, where each chromosome is a binary number representing a set of encoded PEQ parameters, as shown in step 710. A fitness-based filter initialization can be performed, as shown in step 712. New frequency bands can be incrementally added, as shown in step 714. After the performance metric or termination criteria has been satisfied, a set of parameterized EQ band parameters can be returned or obtained from the algorithm, where the set of parameterized EQ band parameters closely or most closely match the desired EQ response and are adapted to the particular room, as shown in step 716.

[0044] Sub-band method: Another aspect of the present disclosure relates to a method / algorithm for adjusting the EQ response of an audio system across the bandwidth of a system microphone while maintaining the factory calibrated EQ response across the remaining bandwidth of the system. For such a method, the details of the individual steps are described and shown below with respect to Figures 8-10 and details of the corresponding higher level algorithm are provided below with respect to Figure 11

[0045] As previously mentioned, the audio characteristics of a television can be factory tuned to provide optimal performance for a nominally standardized room, but when placed in a home or other building, the acoustic environment of any given real world room in which the television is used can alter the spectral characteristics and sound clarity of the television. Tuning the television to a particular room can greatly improve its performance.

[0046] ​When a user (e.g., a "consumer" or "purchaser") performs a home tuning of a television, the remote control or microphone available in the television typically has a narrower bandwidth than the television loudspeakers. Thus, the full bandwidth of the loudspeakers cannot be measured by the microphone. In this case, it is desirable or advantageous to tune only the EQ bandwidth that can be recorded by the microphone. As described below, embodiments of the present disclosure operate according to an algorithm (including variants) that adjusts the EQ of the audio system over the bandwidth of the microphone(s), while maintaining the factory calibrated EQ response over the remaining bandwidth. Thus, the same equalizer that was originally tuned at the factory can share the optimized loudspeaker-to-room (speaker / room) response EQ for use in the home without the need for additional filters specifically used for home adjustment.

[0047] Thus, for exemplary embodiments, room-specific tuning of an audio system (e.g., a television) can be accomplished by following a guided procedure. Such a procedure can use the microphone built into the television, remote control, or smartphone to record an excitation signal played through the television loudspeakers. For the remote control or smartphone, the consumer preferably places the remote control at one or more listener positions to perform the recording. Once the frequency response of the loudspeakers is obtained in this manner, the response can be used to derive an ideal EQ response that equalizes the overall response of the audio system (e.g., the television) to match a target response. Since only the measured frequency response portion over the microphone bandwidth is available (in this case), the ideal EQ response derived from this measurement is spliced with the factory EQ response of the television to create a fully idealized EQ response of the system tailored or adapted to the specific room.

[0048] An initial step before splicing is to match the levels of the two responses. The absolute level of the measured response will vary significantly depending on the microphone position as well as variations in microphone sensitivity and loudspeaker volume. To compensate, the average level of the ideal EQ response over the microphone bandwidth is shifted to match the average level of the factory EQ response over the same bandwidth.

[0049] The spliced response over the microphone bandwidth is equivalent to (or substantially so) the ideal EQ response derived from the home (in-room) measurement. Just outside the microphone bandwidth are two transition regions, of half-octave or about so in width. In calculating the ideal equalizer response in these regions, the spliced response transitions (e.g., linearly) from the measured response to the factory EQ response. Figure 8 FIG. 8 is a graph 800 illustrating an example of a spliced target response according to exemplary embodiments of the present disclosure.

[0050] EQ.1:

[0051]

[0052] Eq. 1: Half octave solution where H1 is the response of the splice in the lower transition region, fi is the lower end of the microphone bandwidth, H meas is the measured ideal EQ response, H fac is the factory calibrated EQ response.

[0053] EQ. 2:

[0054]

[0055] Eq. 2: Half octave solution where H2 is the response of the splice in the upper transition region, f2 is the upper end of the microphone bandwidth, H meas is the measured ideal EQ response, H fac is the factory calibrated EQ response.

[0056] Figure 9 is a plot 900 showing an example speaker response with a steep roll-off for an 8 kHz sampling rate in accordance with example embodiments of the present disclosure. In some cases, the response of a microphone can drop off sharply outside of its nominal bandwidth. This can occur because the recording is being made at a low sampling rate, as Figure 9 indicated. This can cause the ideal EQ response in the transition region to have a sharp spike, making the transition region have an undesirable response when using the above method.

[0057] For this case, an alternative formula (Eq. 3) is preferably used for the upper transition region. The ideal EQ response value at the edge of the microphone bandwidth (H meas (f2)) is used as H meas contribution to the linear transition across the region, rather than H(f). This formula provides a smoother transition region and eliminates the contribution of a large spike in the ideal EQ response from a steep roll-off in the speaker response. Figure 10 is a plot 1000 showing an example of a modified upper transition region for a steep roll-off in accordance with example embodiments of the present disclosure.

[0058] EQ. 3:

[0059]

[0060] Eq. 3: Half octave solution where H 2mod is the response of the modified splice in the upper transition region, f2 is the upper end of the microphone bandwidth, H meas is the measured ideal EQ response, H fac is the factory calibrated EQ response.

[0061] Once the measured response and factory response are spliced together, the automated EQ algorithm in accordance with the present disclosure (e.g., as outlined above forFigures 1-7 The genetic based algorithm described can continue or be used to find the optimal solution for the PEQ parameters. For example, the initial values of the PEQ can be the factory PEQ parameters or can be chosen randomly. The result is that the speaker response of the television is tuned to the room over the bandwidth of the microphone, but the factory EQ response is maintained outside of that bandwidth.

[0062] Figure 11 is a block diagram of an exemplary method 1100 of adjusting the EQ response of an audio system over the bandwidth of a system microphone while maintaining the factory calibrated EQ response over the remaining bandwidth of the system. The method 1100, which can also be considered an algorithm, includes a step in which the measured ideal response of a sub-band (e.g., corresponding to the system microphone) can be spliced into the already pre-calibrated response of the audio system, as shown in step 1102. A transition region can be defined on either side of the spliced response, as shown in step 1104. Statistical techniques can be used to match the level of the spliced response to the already calibrated response, as shown in step 1106. A genetic algorithm can be used to determine the optimal solution for the PEQ parameters of the combination (spliced response inserted into the initial) of the audio system for a particular room, as shown in step 1108.

[0063] Figure 12 is a schematic diagram of an exemplary computer system 1200 that can perform all or at least a portion of the processing, such as the steps in the algorithms and methods described herein and / or the solving of equations EQ. 1-3. The computer system 1200 includes a processor 1202, a volatile memory 1204, a non-volatile memory 1206 (e.g., hard disk), an output device 1207, and a user input or interface (UI) 1208, such as a graphical user interface (GUI), mouse, keyboard, display, or any common user interface, etc. The non-volatile memory (non-transitory storage medium) 1206 stores computer instructions 1212 (a.k.a., machine-readable instructions or computer-readable instructions), such software (computer program product), an operating system 1216, and data 1218. In one example, the computer instructions 1212 are executed by the processor 1202 (from the volatile memory 1204) outside of the volatile memory 1204. In one embodiment, an article of manufacture 1220 (e.g., a storage device or medium, such as a hard disk, optical disk, magnetic storage tape, optical storage tape, flash drive, etc.) includes or stores the non-transitory computer-readable instructions.

[0064] The processes can be implemented in hardware, software, or a combination of both. The processes can be implemented in computer programs executing on programmable computers / machines that each includes a processor, a storage medium or other article of manufacture that is readable by the processor (including volatile and non-volatile memory and / or storage elements), and optionally at least one input device, and at least one output device. Program code can be applied to data entered using an input device or from a machine storage medium, to perform the processes described herein and to generate output information.

[0065] The system 1200 can perform the processes at least partially by a computer program product, e.g., in a machine-readable storage medium, for execution by, or to control the operation of, data processing apparatus (e.g., a programmable processor, a computer, or multiple computers). Each such program can be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, the programs can be implemented in assembly or machine language. The language can be a compiled or interpreted language, and can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. The computer program can be deployed to be executed on one computer or on multiple computers at one site or distributed across multiple sites and interconnected by a communication network. The computer program can be stored on a storage medium or device (e.g., CD-ROM, hard disk, or magnetic diskette) readable by a general or special purpose programmable computer, for configuring and operating the computer when the storage medium or device is read by the computer to cause the computer to perform the procedures described herein. The processes can also be implemented as a machine-readable storage medium configured to have computer programs stored therein, wherein the instructions in the computer programs, when executed, cause the computer to operate.

[0066] The processes can be performed by one or more programmable processors executing one or more computer programs to perform the functions of the system. All or part of the system can also be implemented as, special purpose logic circuitry (e.g., an FPGA (field programmable gate array) and / or an ASIC (application-specific integrated circuit)).

[0067] Example Embodiments: Example embodiments are described in the numbered clauses below.

[0068] Clause 1 : A system to adapt IIR filters of an audio equalizer (EQ) to a particular room, the system comprising: a memory including computer executable instructions; and a processor coupled to the memory and operable to execute the computer executable instructions, the computer executable instructions causing the processor to perform operations comprising: (a) obtaining a system impulse response of a particular audio system of a particular room; (b) converting the impulse response to a frequency response; (c) comparing the frequency response to a target frequency response of the audio system; (d) deriving an ideal EQ response of the system based on the comparison of the frequency response to the target frequency response; (e) implementing a genetic based adaptation algorithm creating a population of chromosomes, where each chromosome is a binary number representing a set of encoded PEQ parameters; (f) performing a fitness based filter initialization; (g) incrementally introducing new frequency bands; and (h) producing a set of parameterized EQ band parameters after performance metrics have been satisfied, wherein the set of parameterized EQ band parameters are adapted to the particular room.

[0069] Clause 2: The system of clause 1, wherein performing a fitness based filter initialization comprises implementing a fitness calculation based on weighted frequency bands.

[0070] Clause 3: The system of clause 1, wherein performing a fitness based filter initialization comprises filter initialization based on high Q implementation at the maximum current error peak.

[0071] Clause 4: The system of clause 1, wherein the processor further performs operations comprising using Gray coding of PEQ parameters for genetic evolution.

[0072] Clause 5: A method for adapting measured sub-bands of an EQ response to a calibrated EQ response of an audio equalizer (EQ) of an audio system of a particular room, the method comprising: (a) creating an EQ target by splicing the ideal response of the measured sub-bands to an already pre-calibrated response; (b) defining a transition region on either side of the spliced response; and (c) matching the level of the spliced response to the calibrated response using a statistical technique.

[0073] Clause 6: The method of clause 5, further comprising using a genetic based adaptation algorithm.

[0074] Clause 7: The method of clause 5, wherein the statistical technique is a mean calculation.

[0075] Clause 8: The method of clause 5, wherein in the transition region calculation, there is a linear change between the measured response and the calculated response.

[0076] Clause 9: The method of clause 5, wherein the measured response used in the transition region calculation is a constant value to avoid spikes in the stitched response caused by out-of-band roll-off in the measurement system.

[0077] Clause 10: The method of clause 5, wherein the width of the transition region is one-half octave.

[0078] Clause 11 : A method of adapting an IIR filter of an audio equalizer (EQ) to a specific room, the method comprising: (a) acquiring a system impulse response; (b) converting the impulse response to a frequency response; (c) implementing a genetic-based adaptation algorithm creating a population of chromosomes, where each chromosome is a binary number representing a set of encoded PEQ parameters; (d) implementing a non-random placement of initial filters for the genetic-based adaptation algorithm; (e) incrementally introducing new frequency bands; and (f) creating the EQ target by stitching the ideal response of the measured sub-bands to a response that has been pre-calibrated.

[0079] Clause 12: The method of clause 11, further comprising defining a transition region on either side of the stitched response.

[0080] Clause 13: The method of clause 11, further comprising using statistical techniques to match the level of the stitched measured response to the pre-calibrated response.

[0081] Clause 14: A computer-readable non-transitory storage medium (or computer program product) comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method of clause 5 and / or as recited in any one of clauses 1-13.

[0082] Clause 15: A computer-readable non-transitory storage medium (or computer program product) comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method of clause 11 and / or as recited in any one of clauses 1-13.

[0083] Accordingly, embodiments of the inventive subject matter can provide advantages over the prior art. These advantages can include, but are not limited to, adapting an off-the-shelf audio system (such as in a high-end television) to a given specific room or other physical location, presenting a specific or unique auditory environment with a set of acoustic characteristics.

[0084] Various embodiments of the claimed concept, system, device, structure, and technology are described above with reference to the associated drawings. Alternative embodiments can be devised without departing from the scope of the described concept, system, device, structure, and technology. It is noted that various connections and positional relationships (e.g., above, below, right, left, rear, front, rearward, forward, upward, downward, etc.) can be described in the drawings and in the specification. These connections and / or positional relationships can be direct or indirect, and the described concept, system, device, structure, and technology are not intended to be limited to direct or indirect connections or positional relationships. Accordingly, the coupling between entities can be direct or indirect, and the positional relationships can be direct or indirect positional relationships.

[0085] As an example of an indirect positional relationship, positioning element “A” above element “B” can include instances where one or more intermediate elements (e.g., element “C”) are between element “A” and element “B” as long as the relevant properties and functions of elements “A” and “B” are not significantly altered by the intermediate element(s).

[0086] Furthermore, the following definitions and abbreviations are to be used for the interpretation of the claims and the specification. The terms “comprise,” “comprises,” “comprising,” “include,” “includes,” “including,” “has,” “have,” “having,” “contain,” “contains,” or “containing,” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0087] Further, the term “exemplary” is intended to mean “serving as an example, instance or illustration.” Any embodiment or design described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments or designs. The terms “at least one” and “one or more” are intended to encompass any integer number greater than or equal to one, i.e., one, two, three, four, etc. The terms “plurality” and “multiple” are intended to encompass any integer number greater than or equal to one, i.e., one, two, three, four, etc. The term “connected” can include “directly connected” and “indirectly connected” via one or more other devices or components.

[0088] References in the specification to “an embodiment,” “one embodiment,” “a

[0089] Relative or positional terms, including but not limited to, the terms “upper,” “lower,” “right,” “left,” “vertical,” “horizontal,” “top,” “bottom,” and derivatives thereof, relate to the orientation of the described structures and methods as oriented in the drawings. The terms “over,” “above,” “on,” “positioned on” or “positioned above” mean that the first element (e.g., a first structure) is present on the second element (e.g., a second structure), where intervening elements (e.g., interface structures) can be present between the first element and the second element. The term “direct contact” means that a first element (e.g., a first structure) and a second element (e.g., a second structure) are connected without any intermediary elements.

[0090] The use of ordinal terms such as “first,” “second,” “third,” etc. in the claims to modify a claim element does not imply any priority, precedence or order of one claim element over another claim element, but is used merely as labels to distinguish one claim element from another.

[0091] The terms “about” and “approximately” can be used to indicate that in some embodiments, the value is within ±20% of the target value, in some embodiments, within ±10% of the target value, in some embodiments, within ±5% of the target value, and in some embodiments, within ±2% of the target value. The terms “about” and “approximately” can include the target value. The term “substantially equal” can be used to refer to values that are within ±20% of each other, in some embodiments, within ±10% of each other, in some embodiments, within ±5% of each other, and in some embodiments, within ±2% of each other.

[0092] The term“substantially” can be used to refer to values within ±20% of a comparative measure in some embodiments, within ±10% in some embodiments, within ±5% in some embodiments, and within ±2% in some embodiments. For example, a first direction that is“substantially” perpendicular to a second direction can refer to a first direction that is within ±20% of a 90° angle to the second direction in some embodiments, within ±10% of a 90° angle to the second direction in some embodiments, within ±5% of a 90° angle to the second direction in some embodiments, and within ±2% of a 90° angle to the second direction in some embodiments.

[0093] The disclosed subject matter is not limited to the details of construction and the arrangement of components set forth in the following description and illustrated in the drawings. The disclosed subject matter is capable of other embodiments and of being practiced and carried out in various ways.

[0094] Furthermore, the phraseology and terminology used by the disclosure are for the purpose of description and should not be regarded as limiting. Thus, the concepts underlying the present disclosure can be readily used as a basis for the designing of other structures, methods and systems for carrying out the several purposes of the disclosed subject matter. Accordingly, the claims should be regarded as including such equivalent constructions insofar they do not depart from the spirit and scope of the disclosed subject matter.

[0095] While the disclosed subject matter has been described and illustrated in the foregoing exemplary embodiments, the present disclosure is merely illustrative of the many possible implementations of the disclosed subject matter. Accordingly, numerous other modifications and changes can be made to the disclosed subject matter without departing from the spirit and scope of the disclosed subject matter.

[0096] The scope of the present patent should therefore not be limited to the described embodiments, but should only be limited by the spirit and scope of the appended claims.

[0097] All publications and references cited in this patent are expressly incorporated by reference in their entirety.

Claims

1. A method for adapting a measured sub-band of an EQ response to a pre-calibrated EQ response of an audio equalizer, EQ, of an audio system of a specific room, the method comprising: a. creating an EQ target by splicing an ideal response of a measured sub-band into the pre-calibrated EQ response, thereby forming a spliced response; b. defining a transition area on both sides of the spliced response; and c. matching a level of the spliced response to the pre-calibrated EQ response using a statistical technique.

2. The method of claim 1, further comprising calculating parametric EQ parameters of the EQ of the audio system using a genetic-based adaptation algorithm. The statistical technique comprises a mean calculation.

3. The method of claim 1, wherein, The defining a transition area on both sides of the spliced response comprises providing a linear variation between the measured sub-band of the EQ response and the pre-calibrated EQ response.

4. The method of claim 1, wherein, The measured sub-band of the EQ response for defining a transition area comprises a constant value to avoid a spike in the spliced response.

5. The method of claim 1, wherein, The width of the transition area is one octave.

6. The method of claim 1, wherein, 7. A computer-readable non-transitory storage medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method of claim 1.

8. A system for adapting a measured sub-band of an EQ response to a pre-calibrated EQ response of an audio equalizer, EQ, of an audio system of a specific room, the system comprising: a memory comprising computer executable instructions; and a processor coupled to the memory and operative to execute the computer executable instructions, the computer executable instructions causing the processor to perform the steps of the method of claim 1. ​ ​

Images