CN110213708B - Nonlinear measurement and tone quality tuning system of loudspeaker system - Google Patents

Abstract

The invention relates to a nonlinear distortion measurement and tone quality tuning system of a loudspeaker system, which explores the commonality of harmonic distortion characteristics of the loudspeaker system and harmonic characteristics of instrumental music/voice sound between a physical level and an auditory psychological level, and carries out real-time nonlinear pre-regulation and control on an input audio stream signal of the loudspeaker system through an innovative audio signal processing algorithm according to the theoretical basis of sound harmony so as to tune between the harmonic characteristics of the input audio signal and the nonlinear characteristics of the loudspeaker system, realize the harmony change of the output sound signal of the loudspeaker system and further improve the pleasant feeling of the sound played by the loudspeaker system. The invention realizes the non-linear tone tuning of the loudspeaker system on the premise of not changing the physical structure and the process of the loudspeaker system, thereby not only controlling the manufacturing cost of the loudspeaker, but also effectively improving the distorted listening feeling of the loudspeaker system, and simultaneously, fully exerting the flexibility and the multifunctionality of the digital audio signal processing platform.

Description

Nonlinear measurement and tone quality tuning system of loudspeaker system

Technical Field

The invention relates to the technical field of nonlinear distortion measurement and tone quality auditory sense compensation innovation of a sound system, in particular to a nonlinear distortion measurement and tone quality tuning system of a loudspeaker system.

Background

As shown in fig. 1, low-frequency and high-frequency nonlinear distortion of a speaker system is a very common phenomenon, which is also a substantial cause of poor sound quality. The traditional loudspeaker sound quality improving means is to improve the material, structure and process of the loudspeaker, but the method is limited by other technical fields and economic cost to a certain extent. With the widespread application of DSP technology, it is a necessary trend to perform speaker sound quality compensation by means of audio signal processing.

At present, the discussion about the nonlinear distortion problem of the loudspeaker is mostly in the theoretical research level, practical application solutions and related products are very rare, and common loudspeaker nonlinear distortion digital compensation techniques can be summarized into the following three types:

(1) predistortion compensation based on Volterra modeling;

the method needs to firstly obtain an inverse system of an N-order Volterra model of the loudspeaker system, is used for constructing an N-order predistortion filter, is connected in series to an input loop of the system to perform predistortion treatment on an input signal and then feeds the input signal to an amplifier, so that an output signal of the loudspeaker system does not contain 2-N-order nonlinear distortion components, and thus nonlinear predistortion compensation of the loudspeaker system is realized. According to the Weierstrass criterion, the convergence of the Volterra series is only guaranteed when the input signal to the system is sufficiently small, so that the Volterra inverse filter is always inserted in front of the power amplifier. In addition, to obtain accurate modeling of the nonlinearity of the speaker system, it is necessary to estimate a sufficient number of Volterra orders to achieve the modeling, and a high-order Volterra filter is difficult to achieve in practical application, so it is usually assumed that the system to be tested can be approximated by the first two or three order Volterra kernels, as shown in fig. 1, that is, assuming that the system is in a weak nonlinear situation, so the practical application range of the Volterra order model is greatly limited.

(2) A method for compensating Klippel mirror image filtering;

the filter structure describing the transmission characteristic of a loudspeaker system under test is derived from a nonlinear differential equation of the loudspeaker system under test, and a filter system with a nonlinear characteristic opposite to that of the loudspeaker system is created to compensate for the nonlinear distortion of the loudspeaker. The structure of this filter is very closely related to a specific loudspeaker system, and it has a correction subsystem, which can be designed according to the mirror symmetry structure of the non-linear elements of the equivalent circuit of the loudspeaker system, so called "mirror filter", as shown in fig. 2. This filter is made up of a minimum number of linear and non-linear subsystems and can also be implemented by digital signal processing. The accuracy of the simulation of the loudspeaker characteristic is limited by the measurement accuracy. The image filter has the advantage that the image filter has certain compensation effect on large signals, small signals and high-order distortion. Compared with the previous compensation method using a Volterra series, the method has the advantages of fewer used elements and lower cost.

(3) An adaptive inverse filter compensation method;

the two compensation methods described above are both feedforward compensation, and neither of them can solve the problem of inconsistency between the actual speaker system and its model, and the parameter variation of the speaker under the action of heat generation, etc., and these problems can only be solved by adaptive techniques. The adaptive system adjusts the system configuration according to an algorithm based on the difference between the expected output signal and the actual output signal. Klippel (1998) proposes a nonlinear adaptive control method for a loudspeaker, which is shown in fig. 3, and divides an input signal into two paths: one path reaches the comparator through the controlled system (loudspeaker, microphone and external noise) and the inverse model thereof; the other path is sent to a comparator after time delay to obtain an error. The error is used to correct the inverse model and the relevant parameters of the controller. When the desired value of the error function reaches its Least Mean Square value, the system is considered to have achieved optimal control, which is the Least Mean Square (LMS) method in the adaptive filtering algorithm.

In the professional audio field, besides the recording and monitoring, the human auditory sense is required to be pleasant and good, and in the consumer audio field, the requirement for extreme reality and the original sound restoring degree are often unrealistic. In order to make the sound good, the pleasure of the sound is maintained or increased to make it have the characteristics of a musical tone. The most prominent expression form of musical tone features is the musical harmony, which is expressed as the regularity of harmonic trains. The harmonicity and the pleasure of the tones can be regulated and controlled through the modulation of the tone harmonic rule, so that the tone quality of the sound is changed. In sound engineering practice, the skilled person has become aware of this problem. For example, by changing the structure of the speaker so that even harmonic distortion is larger than odd harmonic distortion, the sound quality can be improved without reducing the total harmonic distortion, which is the principle that the harmonics of even harmonics are higher than odd harmonics.

Musical tone harmony is the most classical subject of the sound harmony problem and is often associated with auditory pleasure. Summary was made based on the conclusions related to the study of the acoustic perception of musical instruments in musical acoustics, as shown in table 1. The regularity of the harmonic structure of a musical tone is a main factor affecting the auditory sensation of the musical tone, but there is no direct correlation between the regularity and the auditory pleasure. Since the pleasure itself reflects the degree of satisfaction of a person with the listening sensation of a certain sound, generally, in the case where the listener is unfamiliar with a certain sound source, the pleasure is the most effective index for the evaluation of the overall sound quality, and the evaluation of the listening sensation of harmony with the musical sound needs to be established on a certain cognitive basis. Therefore, musical tone harmony is the basis of the sense of audibility, and harmony is not equivalent to the degree of audibility for different listeners.

TABLE 1 correlation table of sound physical elements and auditory sense

Disclosure of Invention

The invention designs a nonlinear measurement and tone quality tuning system of a loudspeaker system, which solves the difficulty and problem of the nonlinear distortion compensation technology of the traditional loudspeaker system.

In order to solve the technical problems, the invention adopts the following scheme:

a non-linear measuring and tone quality tuning system of a loudspeaker system comprises a non-linear distortion and tone quality evaluating module and a non-linear tone quality tuning module; the nonlinear distortion and tone quality evaluation module is used for generating a nonlinear distortion test signal, sending and receiving the test signal, analyzing the frequency spectrum of the test signal, analyzing and displaying the result of nonlinear distortion evaluation and storing data; and the nonlinear tone tuning module performs tone tuning mode configuration, actual audio signal input, nonlinear tone tuning real-time processing and loudspeaker system output according to the nonlinear distortion evaluation result.

Preferably, the loudspeaker system nonlinear distortion and tone quality evaluating module comprises a test signal generating unit and a harmonic spectrum calculating unit; the test signal generating unit successively sends harmonic composite sound signals to the loudspeaker system, harmonic spectrum calculation is carried out on the harmonic composite sound collected by the microphone through the harmonic spectrum calculating unit, and a corresponding harmonic modulation data table is generated and sent to the nonlinear tone tuning module.

Preferably, the fundamental frequency, the number of harmonics, the phase distribution mode and the duration of the harmonic composite tone test signal can be customized; alternatively, the overall spectral characteristics of the harmonic composite tone test signal closely approximate the actual musical and speech-like tones and are also structurally similar to musical instruments from a fine harmonic.

Preferably, the overall frequency range of the harmonic composite tone signal is 22Hz-20kHz, the fundamental frequency values are set to 7 groups according to the pitch range of common musical instruments, each group includes 12 different fundamental frequency values, the frequency value corresponds to 12 semitone pitches in an octave range in music, the pitch names in music are represented and the values of the fundamental frequency values are shown in table 1:

pitch of wind	C	C#	D	D#	E	F	F#	G	G#	A	A#	B
													Fundamental frequency/Hz	65	69	73	78	82	87	92	98	104	110	117	123
Pitch of wind	c	c#	d	d#	e	f	f#	g	g#	a	a#	b
													Fundamental frequency/Hz	131	139	147	156	165	175	185	196	208	220	233	247
Pitch of wind	c1	c1#	d1	d1#	e1	f1	f1#	g1	g1#	a1	a1#	b1
													Fundamental frequency/Hz	262	227	294	313	331	349	369	391	415	440	471	494
Pitch of wind	c2	c2#	d2	d2#	e2	f2	f2#	g2	g2#	a2	a2#	b2
													Fundamental frequency/Hz	523	554	587	622	659	698	740	783	831	880	932	988
Pitch of wind	c3	c3#	d3	d3#	e3	f3	f3#	g3	g3#	a3	a3#	b3
													Fundamental frequency/Hz	1047	1109	1175	1245	1319	1397	1480	1568	1661	1760	1865	1976
Pitch of wind	c4	c4#	d4	d4#	e4	f4	f4#	g4	g4#	a4	a4#	b4
													Fundamental frequency/Hz	2093	2218	2349	2489	2637	2794	2960	3136	3322	3520	3729	3951
Pitch of wind	c5	c5#	d5	d5#	e5	f5	f5#	g5	g5#	a5	a5#	b5
													Fundamental frequency/Hz	4186	4440	4499	4978	5474	5588	5920	6272	6645	7040	7459	7902

TABLE 1 fundamental frequency evaluation of harmonic composite tone signal for nonlinear distortion measurement

The relative phase relation among the harmonic components has three forms of zero phase, random phase and cosine phase; the harmonic number of the harmonic composite sound is adjustable within the range of 6-13; the harmonic composite sound time length is adjustable within the range of 1s-10 s.

Preferably, the nonlinear tone tuning module includes a short-time fourier transform analysis unit, a harmonic spectrum modification unit, and an inverse fourier transform unit; after an input audio signal enters the nonlinear tone tuning module, the modulated audio signal is output after corresponding real-time processing calculation; the real-time processing and calculating process comprises a tone quality tuning mode that an input audio signal is sequentially synthesized by the short-time Fourier transform analysis unit, the harmonic frequency spectrum correction unit and the inverse Fourier transform unit; and the harmonic modulation data table output by the nonlinear distortion and tone quality evaluation module and the input audio signal are subjected to harmonic modulation in the harmonic spectrum correction unit.

A non-linearity measuring and tone quality tuning method of a speaker system includes the steps of:

the harmonic modulation data table of the loudspeaker system to be tested is constructed in advance according to the harmonic spectrum response mode of the loudspeaker system to be tested on the input harmonic composite sound, the harmonic component in the audio signal is detected before the input end of the loudspeaker system, and the harmonic is input into the loudspeaker system after being pre-modulated through the harmonic modulation data table, so that the non-harmonic component in the output signal of the loudspeaker is eliminated or suppressed, and the harmony of the sound played back by the loudspeaker system is improved.

A non-linearity measuring and tone quality tuning method of a speaker system includes the steps of:

step 1, evaluating nonlinear distortion and tone quality of a loudspeaker system; and 2, tuning the nonlinear tone quality of the loudspeaker system.

Preferably, in step 1, a harmonic complex tone signal is successively sent to a certain speaker system to be tested, the test signal replayed by the speaker system to be tested is synchronously collected by a measuring microphone, and harmonic spectrum calculation is performed on the harmonic complex tone collected by the microphone to generate a corresponding harmonic modulation data table.

Preferably, step 11, according to the user-defined parameter setting, a group of harmonic composite tone digital signals is generated by the software system, the sampling frequency of the digital signals is 48kHz or 96kHz, the quantization bit number is 24bit or 32bit, and the generated test signals are automatically stored locally.

Preferably, after the step 12 and the measurement are started, the harmonic composite sound signals are sent to the speaker system to be tested one by one, and a measurement microphone synchronously receives the harmonic composite sound played by the speaker and automatically stores the received signals locally.

Preferably, in step 13, the received multi-frequency sound digital signal enters fourier spectrum calculation, a complex spectrum of the received signal is obtained through Discrete Fourier Transform (DFT), harmonic frequency and harmonic amplitude are calculated through a harmonic screening filter, and finally the harmonic frequency and harmonic amplitude are compared with a harmonic spectrum component of the input test signal, so that a harmonic modulation data table is generated and automatically stored locally.

Preferably, step 2 comprises the following sub-steps:

step 21, one path of digital audio signal stream enters a nonlinear tone tuning module, and each frame of time domain audio signal is converted to a frequency domain through Short Time Fourier Transform (STFT) so as to obtain each frame of complex frequency spectrum data;

step 22, performing median filtering processing on each frame of frequency spectrum data obtained in the step 21 in two directions of a time axis and a frequency axis respectively to obtain two groups of filter output data respectively, wherein the length of a median filter is 5-15 sampling points which are selectable;

step 23, comparing the spectrum amplitude of the two groups of filter output data obtained in the step 22, wherein the comparison method has two optional modes of absolute value comparison and relative value comparison, and two groups of updated filter output data after regular screening are obtained according to amplitude judgment rules of a steady-state spectrum and a transient-state spectrum, wherein one group of data represents a harmonic component spectrum, and the other group of data represents a transient component spectrum;

step 24, performing peak point detection on the harmonic component frequency spectrum obtained in the step 23 to obtain frequencies and spectral amplitudes corresponding to all peak points and an effective frequency range of the peak points;

step 25, in the peak points obtained in step 24, according to the harmonic characteristic rule of the musical tone signal, determining the peak points belonging to the harmonic components, defining the peak points as the harmonic components, and estimating the frequency, amplitude and phase values corresponding to the harmonics;

step 26, performing corresponding parameter modulation on the harmonic component obtained in the step 25 according to the harmonic modulation data table generated in the loudspeaker system nonlinear distortion and tone quality evaluation in the step 1, so as to obtain harmonic component frequency spectrum data after harmonic modulation;

step 27, adding the modulated harmonic component spectrum data obtained in step 26 and the transient component spectrum obtained in step 23 to obtain modulated total spectrum data;

step 28, performing short-time inverse fourier transform (ISTFT) on the modulated total spectrum data obtained in step 27 to obtain a modulated time domain signal, wherein the short-time window length and the frame shift length of the inverse fourier transform are the same as those in step 21;

and step 29, performing time domain overlap-add synthesis on the time domain signals obtained in the step 27 to obtain final modulated time domain signals, and inputting the final modulated time domain signals to a loudspeaker system.

Preferably, in step 21, the sampling frequency of the digital signal processing is selectable to 48kHz or 96kHz, the quantization bit number is selectable to 24bit or 32bit, the short time window length is selectable to 1024, 2048 and 4096 samples, and the frame shift length is selectable to 128, 256 and 512 samples.

Preferably, the synthesis window in step 29 is 256, 512 and 1024 sample selectable, and the synthesis frame shift length is 128 and 256 sample selectable.

A method for non-linear distortion measurement and tone quality tuning of a loudspeaker system, said method being for eliminating or suppressing non-harmonic components of the loudspeaker system output signal, comprising the steps of:

step 1, generating a backward modulation signal of a non-harmonic frequency component, wherein the amplitude of the backward modulation signal needs to be controlled below a non-linear excitation threshold of a loudspeaker system, so that new distortion cannot be generated when the modulation signal passes through the system;

step 2, regarding the non-harmonic and frequency components in a certain harmonic relation, according to the harmonic distortion characteristics of the loudspeaker system, taking the independent non-harmonic and frequency components as loudspeaker output signals, estimating the input signal mode (such as a single-frequency pure tone) of the loudspeaker output signals, and taking the inverted form of the loudspeaker output signals as modulation signals;

and 3, generating a masking modulation signal capable of masking the off-harmonic components by using the auditory masking effect of human ears for the off-harmonic components which are difficult to eliminate by the inverse modulation signal, and simultaneously ensuring that the masking signal has harmonic correlation with the original input signal.

A non-linear distortion measurement and tone quality tuning method for a loudspeaker system is based on a short-time Fourier transform, frequency spectrum correction and inverse Fourier transform method of an input audio stream, and the signal processing flow comprises the following steps: 1) frame-wise truncating the input audio stream through a short analysis window before the audio signal is fed to the speaker system; 2) carrying out discrete Fourier transform on the audio data in each short-time frame, and detecting a frequency spectrum peak point of the audio data; 3) estimating the frequency, amplitude and phase parameter values of each peak point, and modulating the peak point parameter values according to the nonlinear distortion measurement result of the loudspeaker system to be measured; 4) the audio signals of one frame are synthesized anew by inverse fourier transform and time domain waveform overlap-add and then fed to the speaker system.

Compared with the prior nonlinear tone quality compensation technology, the nonlinear measurement and tone quality tuning system and method of the loudspeaker system have the following beneficial effects:

(1) the method is completely different from the existing compensation principle based on the nonlinear modeling and inverse filtering design of the system to be tested, the theoretical basis of the method is from the internal relevance research of the sound harmonicity perception law and the nonlinear distortion of the loudspeaker system, and the method has innovative theoretical support.

(2) Most of the existing nonlinear distortion measurements utilize a single pure tone signal or a multi-frequency sound signal, and these test signals have a large difference from the frequency spectrum components of the audio signals (mostly human voices and musical instrument voices) in practical applications of speakers, so that the obtained test results cannot well reflect the influence of the speaker nonlinearity on the input audio signals. The invention provides a nonlinear measurement system for harmonic composite sound signals, the frequency spectrum characteristics of test signals of the nonlinear measurement system are designed according to the harmonic structure of common music or speech sound, and a harmonic frequency spectrum comparison method between input and output test signals of a loudspeaker system is provided, so that reference data of nonlinear distortion compensation is obtained.

(3) The existing nonlinear distortion compensation methods are realized by time-domain filters, mainly pay attention to the optimization of the structure and the performance of the filters, and the essence of the existing nonlinear distortion compensation methods is a phase-changing linear processing mode, so that the problem of nonlinear distortion of a loudspeaker system cannot be really solved. The invention provides a nonlinear modulation method aiming at time-varying frequency spectrum analysis, correction and synthesis of an input audio stream signal of a loudspeaker, which can more flexibly control the tuning between the harmonic frequency spectrum of the input audio signal and the nonlinearity of a loudspeaker system, thereby improving the playback tone quality of the loudspeaker system and improving the pleasant feeling of the loudspeaker system.

Drawings

Fig. 1 is a schematic diagram of the nonlinear distortion of a prior art loudspeaker system.

Fig. 2 is a schematic diagram of a prior art second order volterra inverse filter structure.

Fig. 3 is a schematic diagram of a mirror filter for compensating a displacement parameter of a voltage driven speaker in the prior art.

Fig. 4 is a schematic diagram of a nonlinear adaptive control system for a loudspeaker system in the prior art.

Fig. 5 is a schematic diagram of the signal processing flow of the non-linear distortion and sound quality evaluation module of the speaker system according to the present invention.

Fig. 6 is a schematic diagram of a signal processing flow of a non-linear tone quality tuning module of a speaker system according to the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terms "including" and "having," and any variations thereof, in the description and claims of this invention and the above-described drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to the listed steps or elements but may alternatively include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments. The invention is further illustrated below with reference to fig. 5 and 6:

the invention relates to a nonlinear distortion measurement and tone quality tuning system and method of a loudspeaker system, which explores the commonality of harmonic distortion characteristics of the loudspeaker system and harmonic characteristics of instrumental music/voice sound between a physical level and an auditory psychological level, and carries out real-time nonlinear pre-regulation and control on an input audio stream signal of the loudspeaker system through an innovative audio signal processing algorithm according to the theoretical basis of sound harmony so as to tune between the harmonic characteristics of the input audio signal and the nonlinear characteristics of the loudspeaker system, thereby realizing the harmony change of the output sound signal of the loudspeaker system and further improving the pleasant feeling of the sound played by the loudspeaker system. The invention realizes the non-linear tone tuning of the loudspeaker system on the premise of not changing the physical structure and the process of the loudspeaker system, thereby not only controlling the manufacturing cost of the loudspeaker, but also effectively improving the distorted listening feeling of the loudspeaker system, and simultaneously, fully exerting the flexibility and the multifunctionality of the digital audio signal processing platform.

Example 1:

as shown in fig. 5, the module for evaluating nonlinear distortion and sound quality of a speaker system according to the present invention includes generating a nonlinear distortion test signal, sending and receiving the test signal, analyzing a frequency spectrum of the test signal, analyzing and displaying a nonlinear distortion evaluation result, and storing data.

The hardware circuit implementation of the loudspeaker system nonlinear distortion and tone quality evaluation module comprises the following parts:

1.1, input signal: 1 path of Mic input signals enter a nonlinear distortion evaluation module;

1.2, main functions: this module provides a measurement mode for the harmonic complex tone test signal: successively sending harmonic composite sound signals to a certain loudspeaker system to be tested, and carrying out harmonic spectrum calculation on the harmonic composite sound collected by the microphone to generate a corresponding harmonic modulation data table;

1.3, output signal: 1 path of output test signal;

1.4, measurement process: the signal processing flow of the module for performing nonlinear measurement on any speaker system to be tested is shown in fig. 5.

The specific process of the measurement process comprises the following steps:

(1) according to the user-defined parameter setting, a group of harmonic composite tone digital signals are generated by a software system, the sampling frequency of the digital signals is 48kHz or 96kHz, the quantization digit is 24bit or 32bit, and the generated test signals are automatically stored locally;

(2) after the measurement is started, the harmonic wave compound sound signals are sent to a loudspeaker system to be measured one by one, a measurement microphone synchronously receives the harmonic wave compound sound played by the loudspeaker, and the received signals are automatically stored locally;

(3) the received multi-frequency sound digital signal enters Fourier spectrum calculation, a complex spectrum of the received signal is obtained through Discrete Fourier Transform (DFT), harmonic frequency and harmonic amplitude are calculated through a harmonic screening filter, and finally the harmonic frequency and the harmonic amplitude are compared with harmonic spectrum components of the input test signal, so that a harmonic modulation data table is generated and automatically stored locally.

Example 2:

as shown in fig. 6, the hardware circuit implementation of the non-linear tone quality tuning module of the speaker system of the present invention includes the following parts:

2.1, input signal: according to the selection of two functional modes of the module, 1 path of actual audio signals are sent to a tuning sub-module of short-time Fourier transform or a tuning sub-module of dynamic harmonic equalization, and only one of the two can be selected.

2.2, main functions: after the 1 channel of actual audio signals enter the nonlinear tone tuning module, the 1 channel of audio signals are output through corresponding real-time processing calculation. The real-time processing calculation process comprises a tone quality tuning mode based on short-time Fourier transform analysis, harmonic frequency spectrum correction and inverse Fourier transform synthesis of an input audio stream signal.

2.3, output signal: the 1 path output corresponds to the 1 path input of the audio signal after the tone quality tuning processing.

2.4, a treatment process: the processing flow of each audio signal is shown in fig. 6.

The specific process of the processing flow comprises the following steps:

step 21, one path of digital audio signal stream enters a nonlinear tone tuning module, and each frame of time domain audio signal is converted to a frequency domain through Short Time Fourier Transform (STFT) so as to obtain each frame of complex frequency spectrum data; the sampling frequency of the digital signal processing is selectable at 48kHz or 96kHz, the quantization bit number is selectable at 24bit or 32bit, the short time window length is selectable at 1024 samples, 2048 samples and 4096 samples, and the frame shift length is selectable at 128 samples, 256 samples and 512 samples.

And step 22, performing median filtering processing on each frame of frequency spectrum data obtained in the step 21 in two directions of a time axis and a frequency axis respectively to obtain two groups of filter output data respectively, wherein the length of the median filter is 5-15 sampling points which are selectable.

And step 23, comparing the spectrum amplitude of the two groups of filter output data obtained in the step 22, wherein the comparison method has two optional modes of absolute value comparison and relative value comparison, and two groups of updated filter output data subjected to rule screening are obtained according to amplitude judgment rules of the steady-state spectrum and the transient-state spectrum, wherein one group of data represents a harmonic component spectrum, and the other group of data represents a transient component spectrum.

And 24, detecting the peak point of the harmonic component frequency spectrum obtained in the step 23 to obtain the frequency and the spectral amplitude corresponding to all the peak points and the effective frequency range of the peak points.

And step 25, in the peak points obtained in the step 24, according to the harmonic characteristic rule of the musical tone signal, determining the peak points belonging to the harmonic components, defining the peak points as the harmonic components, and estimating the frequency, amplitude and phase values corresponding to the harmonics.

And step 26, performing corresponding parameter modulation on the harmonic component obtained in the step 25 according to the harmonic modulation data table generated in the loudspeaker system nonlinear distortion and tone quality evaluation, so as to obtain harmonic component frequency spectrum data after harmonic modulation.

And 27, adding the modulated harmonic component spectrum data obtained in the step 26 and the transient component spectrum obtained in the step 23 to obtain modulated total spectrum data.

And step 28, performing short-time inverse Fourier transform (ISTFT) on the modulated total spectrum data obtained in the step 27 to obtain a modulated time domain signal, wherein the short-time window length and the frame shift length of the inverse Fourier transform are the same as those in the step 21.

And step 29, performing time domain overlap-add synthesis on the time domain signals obtained in the step 27 to obtain final modulated time domain signals, and inputting the final modulated time domain signals to a loudspeaker system. The synthesis window length is 256, 512 and 1024 sample selectable, and the synthesis frame shift length is 128 and 256 sample selectable.

The method detects finer harmonic frequency, amplitude and phase information of the audio signal in a short time window through the forward/inverse transformation of short-time Fourier transform (STFT), and can arbitrarily modify and resynthesize the harmonic information parameters according to the actual application requirements, thereby having great flexibility. However, the method is based on frame-division operation and processing of audio signal streams, and each frame of audio signal is subjected to Discrete Fourier Transform (DFT), harmonic detection, spectral parameter modification, inverse fourier transform synthesis, and the like.

The non-linear measurement and tone quality tuning system and method for a speaker system according to the embodiments of the present invention are described in detail above, and specific examples are applied herein to illustrate the principles and embodiments of the present invention, and the above description of the embodiments is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims (5)

1. A non-linear measurement and tone quality tuning system for a loudspeaker system, comprising: the system comprises a nonlinear distortion and tone quality evaluation module and a nonlinear tone quality tuning module;

the nonlinear distortion and tone quality evaluation module is used for generating a nonlinear distortion test signal, sending and receiving the test signal, analyzing the frequency spectrum of the test signal, analyzing and displaying the result of nonlinear distortion evaluation and storing data;

the nonlinear tone tuning module performs tone tuning mode configuration, actual audio signal input, nonlinear tone tuning real-time processing and loudspeaker system output according to the nonlinear distortion evaluation result; the loudspeaker system nonlinear distortion and tone quality evaluation module comprises a test signal generation unit and a harmonic spectrum calculation unit; the test signal generating unit successively sends harmonic composite sound signals to the loudspeaker system, harmonic spectrum calculation is carried out on the harmonic composite sound collected by the microphone through the harmonic spectrum calculating unit, and a corresponding harmonic modulation data table is generated and sent to the nonlinear tone tuning module.

2. The non-linearity measuring and tone quality tuning system of a speaker system according to claim 1, wherein: the fundamental frequency, the number of harmonics, the phase distribution mode and the time length of the harmonic composite tone test signal can be customized; alternatively, the overall spectral characteristics of the harmonic composite tone test signal closely approximate the actual musical and speech-like tones and are also structurally similar to musical instruments from a fine harmonic.

3. The non-linearity measuring and tone quality tuning system for a loudspeaker system of claim 2, wherein: the whole frequency range of the harmonic composite tone signal is 22Hz-20kHz, 7 groups of fundamental frequency values are set according to the pitch range of common musical instruments, each group comprises 12 different fundamental frequency values, the frequency value corresponds to 12 semitone pitches in an octave range in music, the name of the pitches in the music is shown in a table 1, and the frequency value of the fundamental frequency is shown in the following table:

pitch of wind C C# D D# E F F# G G# A A# B Fundamental frequency/Hz 65 69 73 78 82 87 92 98 104 110 117 123 Pitch of wind c c# d d# e f f# g g# a a# b Fundamental frequency/Hz 131 139 147 156 165 175 185 196 208 220 233 247 Pitch of wind c1 c1# d1 d1# e1 f1 f1# g1 g1# a1 a1# b1 Fundamental frequency/Hz 262 227 294 313 331 349 369 391 415 440 471 494 Pitch of wind c2 c2# d2 d2# e2 f2 f2# g2 g2# a2 a2# b2 Fundamental frequency/Hz 523 554 587 622 659 698 740 783 831 880 932 988 Pitch of wind c3 c3# d3 d3# e3 f3 f3# g3 g3# a3 a3# b3 Fundamental frequency/Hz 1047 1109 1175 1245 1319 1397 1480 1568 1661 1760 1865 1976 Pitch of wind c4 c4# d4 d4# e4 f4 f4# g4 g4# a4 a4# b4 Fundamental frequency/Hz 2093 2218 2349 2489 2637 2794 2960 3136 3322 3520 3729 3951 Pitch of wind c5 c5# d5 d5# e5 f5 f5# g5 g5# a5 a5# b5 Fundamental frequency/Hz 4186 4440 4499 4978 5474 5588 5920 6272 6645 7040 7459 7902

Table 1 fundamental frequency values of harmonic complex tone signals for nonlinear distortion measurement;

the relative phase relation among the harmonic components has three forms of zero phase, random phase and cosine phase; the harmonic number of the harmonic composite sound is adjustable within the range of 6-13; the harmonic composite sound time length is adjustable within the range of 1s-10 s.

4. A non-linearity measuring and tone quality tuning system for a loudspeaker system according to any one of claims 1-3, wherein: the nonlinear tone tuning module comprises a short-time Fourier transform analysis unit, a harmonic frequency spectrum correction unit and an inverse Fourier transform unit;

after an input audio signal enters the nonlinear tone tuning module, the modulated audio signal is output after corresponding real-time processing calculation; the real-time processing and calculating process comprises a tone quality tuning mode that an input audio signal is sequentially synthesized by the short-time Fourier transform analysis unit, the harmonic frequency spectrum correction unit and the inverse Fourier transform unit; and the harmonic modulation data table output by the nonlinear distortion and tone quality evaluation module and the input audio signal are subjected to harmonic modulation in the harmonic spectrum correction unit.

5. A non-linearity measuring and tone quality tuning method of a speaker system includes the steps of:

the harmonic modulation data table of the loudspeaker system to be tested is constructed in advance according to the harmonic spectrum response mode of the loudspeaker system to be tested on the input harmonic composite sound, the harmonic component in the audio signal is detected before the input end of the loudspeaker system, and the harmonic is input into the loudspeaker system after being pre-modulated through the harmonic modulation data table, so that the non-harmonic component in the output signal of the loudspeaker is eliminated or suppressed, and the harmony of the sound played back by the loudspeaker system is improved.

Images