CN117793608A - Method for generating space reverberation sense - Google Patents
Method for generating space reverberation sense Download PDFInfo
- Publication number
- CN117793608A CN117793608A CN202311840048.9A CN202311840048A CN117793608A CN 117793608 A CN117793608 A CN 117793608A CN 202311840048 A CN202311840048 A CN 202311840048A CN 117793608 A CN117793608 A CN 117793608A
- Authority
- CN
- China
- Prior art keywords
- time
- reverberation
- impulse response
- generating
- sound
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 80
- 230000004044 response Effects 0.000 claims abstract description 126
- 238000004088 simulation Methods 0.000 claims abstract description 30
- 238000012546 transfer Methods 0.000 claims abstract description 19
- 238000010276 construction Methods 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 5
- 238000013179 statistical model Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 40
- 230000000694 effects Effects 0.000 description 19
- 238000010586 diagram Methods 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 238000002592 echocardiography Methods 0.000 description 2
- 230000035807 sensation Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013016 damping Methods 0.000 description 1
- 238000012940 design transfer Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Stereophonic System (AREA)
Abstract
The invention discloses a method for generating a space reverberation sense, and relates to the technical field of cabin acoustics. The method for generating the sense of the spatial reverberation comprises the following steps: establishing a virtual sound source in a simulation space according to the listener position and preset sound image parameters; calculating the direct sound time t between the virtual sound source and the listener position 0 The method comprises the steps of carrying out a first treatment on the surface of the According to the time t of the direct sound 0 And a transfer function h (t) of impulse response in the analog space, generates an impulse response time signal h (t) k ) The method comprises the steps of carrying out a first treatment on the surface of the Calculating echo density value tau according to transfer function h (t), and generating time sequence signal delta (t) with discrete period as echo density value tau k ) The method comprises the steps of carrying out a first treatment on the surface of the According to the impulse response time signal h (t k ) And time-series signal delta (t) k ) Construction of sparsity exponential decay function s (t) k ) The method comprises the steps of carrying out a first treatment on the surface of the According to the sparsity exponential decay function s (t k ) A reverberation response pulse signal is generated. By adopting the technology provided by the invention, low occupation can be realizedThe DSP has good adjustability.
Description
Technical Field
The invention relates to the technical field of cabin acoustics, in particular to a method for generating a space reverberation sense.
Background
With the continuous development of the intelligent automobile industry, a vehicle-mounted information entertainment system is taken as a key component part of an intelligent cabin, wherein cabin acoustics is a core mode of human-computer interaction of the automobile cabin. While reverberation greatly affects the user's experience with audio, it can alter the perception and quality of sound. Often, professional recordings use various techniques to control the reverberation to ensure that the resulting recording has the desired sound quality and environmental feel. But the space of car cabin is narrow and small, and interior trim sound absorbing structure arranges inequality factor for sound is shorter at cabin reverberation time, when the stereo is replayed through cabin sound system, because the factor of reverberation is short, is difficult to reproduce the space sense of recording the audio frequency, and cabin acoustic characteristic can change the tone quality of original audio frequency, and the tone color can be shriveled, not plump on the listening sense.
In order to solve the problem, the existing technology for realizing the reverberation effect in a vehicle mainly combines a cabin multichannel sound system, and uses a sampling reverberation or digital reverberation algorithm to realize the reverberation effect of real acoustic scenes such as a music hall, an opera house, a KTV and the like, so as to make up for the sound field perception space insufficient for reproducing recorded audio by the natural reverberation effect in the vehicle.
The method can not adjust sampling pulse according to cabin acoustic environment and sound system, and meanwhile, the requirement on environment background noise is very strict when the sampling reverberation is on site, and the impulse response needs to be further noise-reduced.
Therefore, in practical application, a mathematical model is built by adopting a digital signal processing algorithm, and the reverberation characteristics of the acoustic environment are simulated by constructing an impulse response sequence by designing direct sound, reflection density, reverberation attenuation rate and the like through a filter system. For implementation of the reverberation model, IIR or FIR filter design transfer functions are generally adopted in filter design, and this method can cause the phase of the signal to be changed, so as to affect the spatial sense and the localization sense of the audio. To mitigate phase distortion, more complex filter structures are typically used, compensation networks are introduced, or the effects of phase distortion are reduced by design sacrificing some of the flatness of the frequency response. The above processing may result in excessive occupation of digital signal processors (Digital Signal Processor, DSP) or other dedicated hardware computing resources.
Disclosure of Invention
The invention provides a method for generating a space reverberation sense, which aims to solve the problem that the resource limitation of an automobile audio system in the prior art is difficult to realize a better reverberation effect.
In order to solve the above technical problems, the present invention provides a method for generating a sense of spatial reverberation, where the method for generating a sense of spatial reverberation includes:
and establishing a virtual sound source in the simulation space according to the listener position and preset sound image parameters.
Calculating a direct sound time t between the virtual sound source and the listener position 0 。
According to the direct sound time t 0 And a transfer function h (t) of impulse response in the simulation space generates an impulse response time signal h (t) k )。
Calculating an echo density value tau according to the transfer function h (t), and generating a time-series signal delta (t) with discrete periods of the echo density value tau k )。
According to the impulse response time signal h (t k ) And the time-series signal delta (t) k ) Construction of sparsity exponential decay function s (t) k )。
According to the sparsity exponential decay function s (t k ) A reverberation response pulse signal is generated.
The technical scheme provided by the invention has the beneficial effects that:
by simulating an impulse response similar to a real acoustic scene (a real acoustic scene like a church, concert hall, etc.), i.e. the time delay characteristics of sound in spatial reflection and the attenuation characteristics of the room impulse response (Room Impulse Response, RIR) are simulated. And establishes a sparsity exponential decay model (the calculation), performs constraint equidistant sparsity on the discrete time sequence of the exponential decay function,generating an impulse response time signal h (t k ) And an echo density value tau-adjustable time-series signal delta (t k ) I.e. the impulse response of the analog space where the reverberation sensation can vary.
The method for generating the reverberation sound effect can realize the reverberation effect of different acoustic spaces (simulation spaces) in a narrow space (such as a cabin) by a user, and the method for generating the reverberation sound effect does not need to use a filter to realize a reverberation model, so that the calculation force of the DSP is reduced, and meanwhile, the phase distortion of the audio sound quality caused by the filter is further reduced.
The method has smaller complexity, so that the method has better reverberation effect, occupies less computational power of the DSP, and has good practicability and adjustability compared with the conventional reverberation generation method.
In some embodiments, the method is based on the direct sound time t 0 And a transfer function h (t) of impulse response in the simulation space generates an impulse response time signal h (t) k ) Comprising:
performing complete sparse representation base h (t) on the transfer function h (t) k )：
Wherein p is 0 For the initial sound source amplitude at the first point in time of the impulse response, λ is the decay rate of the acoustic energy and L is the length of discretized time.
Constructing a sparse vector A:
wherein the length of the sparse vector A is equal to h (t k ) Identical and consisting of elements 0 and 1.
Sparse impulse response vector S (t k ) The method comprises the following steps:
wherein t is 0.ltoreq.t k ≤L。
Further, the echo density value tau is calculated according to the transfer function h (t), and a time series signal delta (t) with discrete period of the echo density value tau is generated k ) Comprising:
the position a of the first non-zero element in the sparse vector A e Is a discrete point, wherein the discrete point is a pre-delay parameter conforming to the impulse response, the number of non-zero elements of the sparse vector A is the echo density value tau, and the amplitude of the values of the non-zero elements of the sparse vector A is the amplitude of the impulse response in a real acoustic scene
By adopting the technical scheme, the echo density value is the number of echoes per second, at present, in order to ensure that the sound is natural after reverberation treatment, the echo density value (1000/s) recommended by Schroeder is generally adopted, but the mode is that sound waves are repeatedly reflected between two parallel wall bodies, and in a real application scene, the reflection of the sound waves is diffusion reflection. The echo density value of the late reverberation can be effectively adjusted by changing the interval of the impulse response sequence, wherein the impulse response in the simulation space is generated by simulating the impulse response in the real acoustic scene by intelligent equipment such as a computer.
Further, the pulse response time signal h (t k ) And the time-series signal delta (t) k ) Construction of sparsity exponential decay function s (t) k ) Comprising:
wherein p is 0 For the initial sound source amplitude at the first point in time of the impulse response, L is the length of the discretized time.
Wherein,
wherein fs is the signal sampling frequency; t is t e Representing the delay time and attenuation coefficient of the early reflected sound in the simulation spaceRT is the reverberation time.
By adopting the technical scheme, the discretized time domain information of the sparse impulse response in the discretized time domain information can be regarded as a discrete time sequence of an exponential decay function and the time delay is D e The period is the sparsity impact function delta (t k ) I.e. sparsity exponential decay function s (t k )。
Wherein when the sampling resolution is 1/fs, the discrete delay time D of the impulse response sequence e The values of n and echo density τ are integers and the length of the discretization time L is related to the sampling frequency fs of the signal.
In some embodiments, the step of generating the sparse representation comprises generating a sparse representation based on the sparse representation k ) Generating a reverberation response pulse signal comprising:
calculating a pre-delay of the simulation space for early reflections within the simulation space i timesAnd generating i sets of impulse responses, wherein the i sets of impulse responses apply different values of reverberation time and different values of echo density values τ; superimposing the i groups of impulse responses in parallel and generating reverberation response pulsesA signal.
By adopting the technical scheme, the echo density of the single impulse response sequence is insufficient, and the frequency spectrum of the sound is periodically changed due to the echo density at equal intervals, so that the sound after the convolution of the audio and the impulse response is severely distorted.
Therefore, different delay times are sampled to generate a plurality of groups of impulse response sequences, and the echo density can be effectively increased after superposition.
Further, the amplitude p of the impulse response for the i sets 0 Multiplying by-1 to obtain amplitude p 0 The impulse response of negative number is superimposed in parallel with the i groups of magnitudes p 0 Impulse response and amplitude p being positive numbers 0 An impulse response that is negative and generates a reverberation response impulse signal.
By adopting the technical scheme, the impulse response with the negative amplitude is increased, namely the negative echo density is obtained to simulate the change of the sound wave phase, so that the more real and natural space reverberation effect can be achieved after the sound wave phase is overlapped.
In some implementations, the calculating a direct sound time t between the virtual sound source and the listener position 0 Comprising:
the listener position r has coordinates (x) r ，y r ，z r ) The method comprises the steps of carrying out a first treatment on the surface of the The coordinates of the virtual sound source s corresponding to the listener position r are recorded as (x) s ，y s ，z s )。
Calculating the direct sound distance d 0 The method comprises the following steps:
calculating the direct sound time t 0 The method comprises the following steps:
where c represents the propagation velocity of the acoustic wave.
Further, the early reflected sound of the virtual sound source in the simulated space is:
the listener position r has coordinates (x) r ，y r ，z r ) The method comprises the steps of carrying out a first treatment on the surface of the The coordinates of the virtual sound source s corresponding to the listener position r are recorded as (x) s ，y s ，z s ) The method comprises the steps of carrying out a first treatment on the surface of the The simulation space is noted as a space of length X, width Y and height Z.
Calculating the distance d of the two or three reflections to the listener's position r ijm The method comprises the following steps:
wherein,
delay time t of the ith early reflected sound of the simulation space e The method comprises the following steps:
where c represents the propagation velocity of the acoustic wave.
By adopting the technical scheme, the direct sound means that the sound source signal directly reaches the human ear without passing through any obstacle, the distance from the sound source to the listening position is the first pulse of the room impulse response, the energy loss is minimum, and the amplitude is maximum. The early reflected sound is that the sound source reaches the listening position after being reflected by the surface of a space wall, a stage and the like for the first time or several times, and the time interval between the direct sound and the first reflected sound is the initial delay.
By establishing the virtual sound source, parameters required for generating the subsequent reverberation can be effectively acquired, and when the parameters are applied to a narrow space (such as in a vehicle-mounted environment), the positions of simulated sound images corresponding to different listeners of the sound field mode of the vehicle-mounted sound system are also different.
In some embodiments, the establishing the virtual sound source in the simulation space according to the listener position and the preset sound image parameter in the real acoustic scene includes:
the listener position comprises a main driving position, a co-driving position and a rear-row position, and the sound image parameters are sound field parameters of an audio module integrated in the DSP chip.
By adopting the technical scheme, for example, in a car machine, a manufacturer can preset a sound field mode aiming at an audio module, wherein corresponding sound image parameters can be set in the sound field mode, so that a certain sound effect is adjusted. And further effectively reduces the difficulty of personalized adjustment of the user.
In some embodiments, the method of generating a sense of spatial reverberation further comprises:
considering the real acoustic scene as a linear time-invariant system, the impulse response within the real acoustic scene is considered as a process in which the energy of the unit impact signal δ (t) gradually decays with time in an exponential function form:
h(t)＝δ(t-τ)+gδ(t-2τ)+g 2 δ(t-3τ)+…
where δ (t) is the dirac function, τ is the echo delay time, and g is the attenuation coefficient.
Simplifying according to a Polack reverberation statistical model:
h(t) 2 ＝E 0 e -λt
wherein,as the energy of the sound source at t=0, λ is the decay rate of the acoustic energy.
Generating a transfer function h (t) of the impulse response in analog space, expressed as:
h(t)＝p 0 e -λ/2t
wherein t is k Is the duration, p 0 The initial sound source amplitude is the first point in time of the impulse response.
With the above technical solution, from the point of view of digital signal processing, the space can be theoretically regarded as a linear time-invariant system (Linear and time invariance system, LTI system), and accordingly, the impulse response in a real acoustic scene can be regarded as a process in which the energy of the unit impact signal gradually decays with time in an exponential function form. And then, the parameters of the model are modulated and designed by establishing a mathematical model (Polack reverberation statistical model), so that the structure of the model can be simplified.
Drawings
For a clearer description of the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly introduced below, it will be obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art, wherein:
FIG. 1 is a flowchart illustrating an embodiment of a method for generating a sense of spatial reverberation according to the present invention;
FIG. 2 is a diagram illustrating an embodiment of an impulse response in a real acoustic scene based on the present invention;
FIG. 3 is a simplified block diagram of an impulse response of simulated reverberation based on a real acoustic scene provided by the present invention;
FIG. 4 is a schematic diagram of a propagation model of an embodiment of the spatial reverberation propagation of a sound field pattern in a vehicle-mounted space according to the present invention;
FIG. 5 is a schematic block diagram of an embodiment of a reverberation model of a method of generating a sense of spatial reverberation according to the present invention;
FIG. 6 is a schematic diagram of impulse response generated by a single sparsity exponential function of an embodiment of a method for generating a sense of spatial reverberation;
FIG. 7 is a schematic diagram of impulse responses generated by parallel connection of 6 sparsity index functions according to an embodiment of a method for generating a sense of spatial reverberation;
FIG. 8 is a sparsity exponentially decaying reverberation model according to one embodiment of a method of generating a sense of spatial reverberation;
fig. 9 is a schematic diagram of a parallel structure of sparsity exponential decay functions according to an embodiment of a method for generating a sense of spatial reverberation.
Detailed Description
The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments, and all other embodiments obtained by those skilled in the art without making any inventive effort based on the embodiments of the present invention are within the scope of protection of the present invention.
In order to facilitate the following description, terms or english letters referred to in the present embodiment are explained below.
Room impulse response, (Room Impulse Response, RIR) characterizes the system response of a room system, and can be used for purposes such as room equalization and calculating room acoustic parameters.
The direct sound, the sound source signal reaches the human ear directly without any obstacle, represents the distance from the sound source to the listening position, is the first pulse of the RIR, and has the least energy loss and the greatest amplitude.
Early reflections sound, the sound source signal is reflected to the listening position after first or several times of reflection on the surfaces of the space wall, the stage and the like.
And (3) initially delaying, namely, the time interval between the direct sound and the first reflected sound.
The late reverberation and the early reflected sound are reflected repeatedly, the density of the late reverberation is increased continuously, and the energy is reduced exponentially.
Linear time invariant system (Linear and time invariance system, LTI system), two keys of LTI system: "Linear" and "time-invariant", which provide the system with three important features: homogeneity, superimposability, time invariance.
Referring to fig. 1, fig. 1 is a flowchart illustrating an embodiment of a method for generating a sense of spatial reverberation provided in the present application.
In some embodiments, the method for generating a sense of spatial reverberation includes:
referring now to fig. 2-3, fig. 2 shows a measured graph of an embodiment of the present application based on impulse responses within a real acoustic scene; fig. 3 shows a simplified block diagram of an impulse response of simulated reverberation based on a real acoustic scene provided herein.
In order to achieve a more realistic effect, the impulse response in a real acoustic scene is obtained, and the sense of realism and sense of reverberation after the reverberation simulation can be greatly improved. Illustratively, the test sound is played by a professional playback device within an acoustic scene (e.g., concert hall, video studio), so that the reverberation characteristics within a real acoustic scene can be obtained by measuring the impulse response of the sound source (e.g., sound) to the sensor.
Illustratively, as shown in connection with fig. 2, the temporal distribution and intensity of the impulse response is closely related to the size, shape of the room. The relationship between the direct sound, the early reflected sound and the late reverberation in real acoustic scenes in time and energy can simulate the reverberation in a space.
From a digital signal processing point of view, a real acoustic scene can theoretically be considered as a linear time-invariant system (Linear and time invariance system, LTI system), and accordingly the impulse response within a real acoustic scene is considered as a process in which the energy of the unit impact signal δ (t) gradually decays with time in an exponential function:
h(t)＝δ(t-τ)+gδ(t-2τ)+g 2 δ(t-3τ)+…
wherein, delta (t) is a dirac function, i.e. ideal impact signal, tau is echo delay time, density between acoustic energy is determined, g is attenuation coefficient, and speed of delta (t) initial energy attenuation is determined, so that g is less than or equal to 1.
As shown in connection with fig. 2 to 3, fig. 2 is simplified to fig. 3 based on three parts (direct sound, early reflected sound, and late reflected sound):
and (3) establishing a mathematical model (Polack reverberation statistic model) to modulate and design parameters, wherein the parameters are simplified into:
h(t) 2 ＝E 0 e -λt
wherein,as the energy of the sound source at t=0, λ is the decay rate of the acoustic energy.
The transfer function h (t) of the impulse response in the simulation space is generated and can be approximated as:
h(t)＝p 0 e -λ/2t
wherein t is k Is the duration, p 0 For the initial sound source amplitude at the first point in time of the impulse response, lambda/2 is the damping constant. Wherein the energy of the impulse response is reduced by an attenuation coefficientIs gradually decayed. RT is the reverberation time, 13.82 is the log of the energy attenuation 60dB, a=log (10≡6).
In some embodiments, a method of generating a sense of spatial reverberation includes:
step S10, a virtual sound source in the simulation space is established according to the listener position and preset sound image parameters.
In this embodiment, when the application scene is a cabin, the listener position includes a main driving position, a co-driving position, and a rear-row position, and the sound image parameter is a sound field parameter of an audio module integrated in the DSP chip.
For example, in a car set, a manufacturer can preset a sound field mode for an audio module, wherein corresponding sound image parameters can be set in the sound field mode, so that a certain sound effect is adjusted. And further effectively reduces the difficulty of personalized adjustment of the user. The method and the device are not limited to a narrow space, and can be applied to an audio-visual room and the like in a similar cabin space, so that the generation of the sense of the spatial reverberation can be realized.
And establishing a virtual sound source to acquire parameter values required by subsequent calculation. Wherein the virtual sound source is mirror-symmetrical with the actual sound source, the virtual sound source is located in front of each listener, or the virtual sound source is set based on the sound field mode.
Referring to fig. 4, fig. 4 is a schematic diagram of a propagation model of an embodiment of the spatial reverberation propagation of the sound field pattern in the vehicle space provided in the present application.
Step S20, calculating the direct sound time t between the virtual sound source and the listener position 0 。
For example, as shown in fig. 4, a main driving position in the cabin space is taken as an example, wherein in the acoustic mode, a main driving mode can be adopted, and a simulated sound image correspondingly generated is taken as a subsequent virtual sound source s.
The listener position r has coordinates (x) r ，y r ，z r ) The method comprises the steps of carrying out a first treatment on the surface of the The coordinates of the virtual sound source s corresponding to the listener position r are recorded as (x s ，y s ，z s ) The method comprises the steps of carrying out a first treatment on the surface of the The analog space is noted as a space of length X, width Y, and height Z.
Calculating the direct sound distance d 0 The method comprises the following steps:
calculating the direct sound time t 0 The method comprises the following steps:
where c represents the propagation velocity of the acoustic wave.
Taking main driving as an example, the position of the main driver is the listener position r, and the direct sound distance d 0 Is the solid line distance between the main driving position and the virtual sound source s. For example, in practical applications, the direct sound distance can also be obtained by directly measuring the distance between the two, with units accurate to CM.
Further, the early reflected sound of the virtual sound source in the simulated space is:
calculating the distance d of the two or three reflections to the listener's position r ijm The method comprises the following steps:
wherein,
delay time t of ith early reflected sound of simulation space e The method comprises the following steps:
where c represents the propagation velocity of the acoustic wave.
That is, the expected delay time is:
the delay time determines the virtual space size, and the longer the delay time is, the larger the simulation space is. The delay time is set by the user by himself considering the reverberation experience in a small space and the influence of the reverberation on the sound quality. In general, when the reverberation time RT of the simulation space to which the subsequent reverberation model calculation is applied is smaller than 1,3s, max [ t ] e ]Less than or equal to 50ms; or, max [ t ] e ]And less than or equal to 80ms. The initial amplitude of the early reflected sound isAmplitude of late reverberant sound of impulse response follows p 0 Gradually decaying at a rate of decay coefficient lambda.
Step S30, according to the direct sound time t 0 And the transfer function h (t) generates an impulse response time signal h (t) with continuous discrete time points in the analog space k )。
The echo density value is how many echoes are per second, at present, in order to ensure that the sound is natural after reverberation treatment, the echo density value (1000/s) recommended by Schroeder is generally adopted, but the mode is to consider that sound waves are repeatedly reflected between two parallel wall bodies, and in a real application scene, the reflection of the sound waves is in diffuse reflection. In practical application, the echo density value of the late reverberation can be effectively adjusted by changing the interval of the impulse response sequence.
The transfer function h (t) generates a continuous signal, so that constraint equidistant sparsification processing is performed on the time domain sequence of the impulse response according to the echo density value. Comprising the following steps:
performing complete sparse expression base h (t) on transfer function h (t) k ) Namely, sparse representation:
wherein p is 0 For the initial sound source amplitude at the first point in time of the impulse response, λ is the decay rate of the acoustic energy and L is the length of discretized time.
Constructing a sparse vector A:
wherein the length of the sparse vector A is equal to h (t k ) Identical and consisting of elements 0 and 1.
Sparse impulse response vector S (t k ) The method comprises the following steps:
wherein t is 0.ltoreq.t k ≤L。
Position a of first non-zero element in sparse vector A e Is a discrete point, wherein the discrete point is a pre-delay parameter of the compliant impulse response, the number of non-zero elements of the sparse vector A is an echo density value tau, and the amplitude of the value of the non-zero element of the sparse vector A is the amplitude of the impulse response in a real acoustic scene
Step S40, calculating echo density value tau according to transfer function h (t), and generating dispersionTime-series signal delta (t) with period of echo density value tau k )：
Wherein,
wherein fs is the signal sampling frequency; t is t e Representing the delay time and attenuation coefficient of early-onset reflected sound in analog space
Sampling resolution is 1/fs, D e Is the discrete delay time of the impulse response sequence. Wherein D is e And τ are integers, L is the length of the discretization time and is related to the sampling frequency fs of the signal. Illustratively, the sampling frequency refers to how much the recording device samples the analog signal in a unit time, and the higher the sampling frequency, the more natural the waveform of the mechanical wave. The sampling rate used in the present application is different from the audio sampling rate of the general playing, for example, the sampling rate is 48000 points.
Step S50, according to the impulse response time signal h (t k ) And time-series signal delta (t) k ) Construction of sparsity exponential decay function s (t) k ). Comprising the following steps:
wherein p is 0 For the initial sound source amplitude of the first time point of the impulse response, L is the length of discretization time, and t is more than or equal to 0 k ≤L。
The discretized time domain information of the sparse impulse response can be regarded as a discrete time sequence of an exponential decay function and the time delay is D e The period is the sparsity impact function delta (t k ) I.e. sparsity exponential decay function s (t k )。
Step S60, according to the sparsity exponential decay function S (t k ) A reverberation response pulse signal is generated.
The impulse response sequence can be obtained from the sparsity exponential decay function, the above-described reverberation processing is performed on the audio, the impulse response signal after the reverberation processing is generated, and the music signal is played by an acoustic device (for example, a vehicle acoustic system).
In the embodiment of the application, the impulse response similar to a real acoustic scene is simulated, namely, the time delay characteristic of sound in space reflection and the attenuation characteristic of RIR are simulated. And establishing a sparsity exponential decay model (namely the calculation), carrying out constraint equidistant sparsity on a discrete time sequence of an exponential decay function, and generating an impulse response time signal h (t) with continuous discrete time points in a simulation space k ) And an echo density value tau-adjustable time-series signal delta (t k ) I.e. the impulse response of the analog space where the reverberation sensation can vary.
The method for generating the reverberation sound effect can realize the reverberation effect of different acoustic spaces (simulation spaces) in a narrow space (such as a cabin) by a user, and the method for generating the reverberation sound effect does not need to use a filter to realize a reverberation model, so that the calculation force of the DSP is reduced, and meanwhile, the phase distortion of the audio sound quality caused by the filter is further reduced.
The method has low complexity, and combines the distribution of vehicle personnel and the sound image parameters corresponding to the sound field mode when the method is applied to the vehicle space. The generated reverberation sound effect is more natural and real.
In addition, the sparse exponential decay reverberation algorithm model with the parameters being simple and adjustable is provided by the method, so that the sparse exponential decay reverberation algorithm model is strong in adjustability, namely, a user can control the adjustable parameters (such as a virtual sound source position and the like) of the model according to personal preference to realize different reverberation perception degrees of audio. And the method has the advantages of realizing better reverberation effect, occupying less calculation power of the DSP, and having good practicability and adjustability compared with the conventional reverberation generation method.
Referring to fig. 5, fig. 5 is a schematic diagram illustrating an embodiment of a reverberation model of a method for generating a sense of spatial reverberation according to the present application.
In some embodiments, the main parameters of the reverberation model include the direct sound, the initial delay time, and the early reflected sound. The design steps of the reverberation model are as follows:
(1) Determining the size and purpose of the simulated acoustic space; illustratively, impulse responses within a real acoustic scene are analyzed based on a smart device (e.g., a computer) and the use of the simulated space (e.g., a user-selected sound field pattern) is determined, as well as a preset reverberation time RT, wherein the reverberation time is preset by sound field parameters of the sound field pattern or is selected by a user's autonomous operation.
(2) Determining the sound image and the listener position corresponding to the current sound field mode; illustratively, step S10 is performed, main parameters such as the direct sound, the initial delay time, and the early reflected sound in the simulation space are acquired, step S20 is performed, and the direct sound time t is calculated 0 And calculating a delay time t e Calculating an initial amplitude p 0 。
(3) Determining an echo density; calculating the echo density value τ, i.e. executing step S40, to generate a time-series signal δ (t) with discrete periods of echo density value τ k )。
(4) Generating a simulated spatial impulse response signal based on the sparsity exponential decay function; based on step S50 and step S60, the listener is enabled to finally hear the reverberated music signal.
Referring to fig. 6 to 7, fig. 6 is a schematic diagram illustrating an impulse response generated by a single sparsity exponential function according to an embodiment of a method for generating a sense of spatial reverberation provided in the present application; FIG. 7 is a schematic diagram of impulse responses generated by parallel connection of 6 sparsity index functions according to one embodiment of a method for generating a sense of spatial reverberation.
In some embodiments, according to thinHydrophobic exponential decay function s (t) k ) Generating a reverberation response pulse signal comprising:
calculating the pre-delay of the simulation space of early reflected sound in the i-th simulation spaceAnd generating i sets of impulse responses, wherein the i sets of impulse responses apply different values of reverberation time and different values of echo density values τ; and superposing i groups of impulse responses in parallel, and generating a reverberation response impulse signal.
In the embodiment of the application, the echo density of the single impulse response sequence is insufficient, and the echo density of equal intervals can cause the frequency spectrum of sound to show periodic variation, so that serious distortion occurs to the sound after the convolution of the audio and the impulse response. Therefore, different delay times are sampled to generate a plurality of groups of impulse response sequences, and the echo density can be effectively increased after superposition. Illustratively, the absolute difference of the reverberation times RT is less than 0.5S, and the difference interval of the echo density values τ is 100.
Further, the amplitude p of the impulse response for group i 0 Multiplying by-1 to obtain amplitude p 0 The impulse response of negative number is superimposed in parallel with the i groups of magnitudes p 0 Impulse response and amplitude p being positive numbers 0 An impulse response that is negative and generates a reverberation response impulse signal.
Illustratively, an impulse response of negative magnitude is added, i.e., a negative echo density is obtained to simulate the change of the acoustic wave phase, so that a more realistic and natural spatial reverberation effect can be achieved after the superposition thereof. In some application scenarios, considering the limitation of calculation power and memory during DSP integration, the value of i is generally 4 to 6 before the value, and in combination with fig. 7, the value of i is 5.
Referring to fig. 8 to 9, fig. 8 illustrates a sparse exponential decay reverberant model of an embodiment of a method for generating a sense of spatial reverberation provided in the present application, and fig. 9 illustrates a schematic diagram of a parallel structure of sparse exponential decay functions of an embodiment of a method for generating a sense of spatial reverberation provided in the present application.
In some embodiments of the present invention, in some embodiments,as shown in connection with FIG. 7, 5 p are employed 0 Is positive and 1 p 0 RIR generated by parallel connection of negative sparsity exponential decay functions can obtain that the amplitude of the finally generated impulse response presents nonlinear exponential decay characteristics, and the echo density is from sparse to dense. I.e. acoustic signal x (n) to acoustic signal y (n).
The foregoing is only the embodiments of the present invention, and therefore, the patent scope of the invention is not limited thereto, and all equivalent structures or equivalent processes using the contents of the present specification and drawings, or direct or indirect application in other related technical fields, should be carried within the protection scope of the present invention.
Claims (10)
1. A method of generating a sense of spatial reverberation, comprising:
establishing a virtual sound source in a simulation space according to the listener position and preset sound image parameters;
calculating a direct sound time t between the virtual sound source and the listener position 0 ；
According to the direct sound time t 0 And a transfer function h (t) of impulse response in the simulation space generates an impulse response time signal h (t) with continuous discrete time points k )；
Calculating an echo density value tau according to the transfer function h (t), and generating a time-series signal delta (t) with discrete periods of the echo density value tau k )；
According to the impulse response time signal h (t k ) And the time-series signal delta (t) k ) Construction of sparsity exponential decay function s (t) k )；
According to the sparsity exponential decay function s (t k ) A reverberation response pulse signal is generated.
2. The method for generating a sense of spatial reverberation according to claim 1, wherein the direct sound time t is based on 0 And the transfer function h (t) of impulse response in the simulation space generates impulse response time signals h (t) with continuous discrete time points k ) Comprising:
performing complete sparse representation base h (t) on the transfer function h (t) k )：
Wherein p is 0 Initial sound source amplitude for the first time point of impulse response, lambda is the attenuation rate of sound energy, and L is the length of discretized time;
constructing a sparse vector A:
wherein the length of the sparse vector A is equal to h (t k ) Identical and consisting of elements 0 and 1;
sparse impulse response vector S (t k ) The method comprises the following steps:
wherein t is 0.ltoreq.t k ≤L。
3. The method of generating a sense of spatial reverberation according to claim 2, wherein the calculating an echo density value τ from the transfer function h (t) and generating a time-series signal δ (t) with a discrete period of the echo density value τ k ) Comprising:
the position a of the first non-zero element in the sparse vector A e Is a discrete point, wherein the discrete point is a pre-delay parameter conforming to the impulse response, the number of non-zero elements of the sparse vector A is the echo density value tau, and the amplitude of the values of the non-zero elements of the sparse vector A is the amplitude of the impulse response in a real acoustic scene
4. A method of generating a sense of spatial reverberation according to claim 3 wherein the signal is derived from the impulse response time signal h (t k ) And the time-series signal delta (t) k ) Construction of sparsity exponential decay function s (t) k ) Comprising:
wherein p is 0 Initial sound source amplitude for the first time point of impulse response, L is the length of discretized time;
wherein,
wherein fs is the signal sampling frequency; t is t e Representing the delay time and attenuation coefficient of the early reflected sound in the simulation spaceRT is the reverberation time.
5. The method of generating a sense of spatial reverberation according to claim 1, wherein the step of generating the sense of spatial reverberation according to the sparsity exponential decay function s (t k ) Generating a reverberation response pulse signal comprising:
calculating i times the modes of early reflected sound in the simulation spacePseudo-spatial pre-delayAnd generating i sets of impulse responses, wherein the i sets of impulse responses apply different values of reverberation time and different values of echo density values τ;
and superposing the i groups of impulse responses in parallel, and generating a reverberation response impulse signal.
6. The method of generating a sense of spatial reverberation as set forth in claim 5 wherein the magnitudes p of the i sets of impulse responses 0 Multiplying by-1 to obtain amplitude p 0 The impulse response of negative number is superimposed in parallel with the i groups of magnitudes p 0 Impulse response and amplitude p being positive numbers 0 An impulse response that is negative and generates a reverberation response impulse signal.
7. The method of generating a sense of spatial reverberation according to claim 1, wherein the calculating a direct sound time t between the virtual sound source and the listener position 0 Comprising:
the listener position r has coordinates (x) r ，y r ，z r )；
The coordinates of the virtual sound source s corresponding to the listener position r are recorded as (x) s ，y s ，z s )；
Calculating the direct sound distance d 0 The method comprises the following steps:
calculating the direct sound time t 0 The method comprises the following steps:
where c represents the propagation velocity of the acoustic wave.
8. The method of generating a sense of spatial reverberation according to claim 1 or 5, wherein the early reflected sounds of the virtual sound sources in the simulated space are:
the listener position r has coordinates (x) r ，y r ，z r )；
The coordinates of the virtual sound source s corresponding to the listener position r are recorded as (x) s ，y s ，z s )；
The simulation space is recorded as a space with length X, width Y and height Z;
calculating the distance d of the two or three reflections to the listener's position r ijm The method comprises the following steps:
wherein,
delay time t of the ith early reflected sound of the simulation space e The method comprises the following steps:
where c represents the propagation velocity of the acoustic wave.
9. The method of generating a sense of spatial reverberation according to claim 1, wherein the creating a virtual sound source in the simulated space from the listener position and a preset sound image parameter comprises:
the listener positions include a primary driving position, a secondary driving position, and a rear-row position; the sound image parameters are sound field parameters of an audio module integrated in the DSP chip.
10. The method of generating a sense of spatial reverberation as set forth in claim 1, wherein the method of generating a sense of spatial reverberation further comprises:
considering a real acoustic scene as a linear time-invariant system, the impulse response within the real acoustic scene is considered as a process in which the energy of the unit impact signal δ (t) gradually decays with time in an exponential function form:
h(t)＝δ(t-τ)+gδ(t-2τ)+g 2 δ(t-3τ)+…
wherein delta (t) is a dirac function, tau is echo delay time, and g is attenuation coefficient;
simplifying according to a Polack reverberation statistical model:
h(t) 2 ＝E 0 e -λt
wherein,is the energy of the sound source at t=0, λ is the decay rate of the acoustic energy;
generating a transfer function h (t) of the impulse response in analog space, expressed as:
h(t)＝p 0 e -λ/2t
wherein t is k Is the duration, p 0 The initial sound source amplitude is the first point in time of the impulse response.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311840048.9A CN117793608A (en) | 2023-12-28 | 2023-12-28 | Method for generating space reverberation sense |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311840048.9A CN117793608A (en) | 2023-12-28 | 2023-12-28 | Method for generating space reverberation sense |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117793608A true CN117793608A (en) | 2024-03-29 |
Family
ID=90398122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311840048.9A Pending CN117793608A (en) | 2023-12-28 | 2023-12-28 | Method for generating space reverberation sense |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117793608A (en) |
-
2023
- 2023-12-28 CN CN202311840048.9A patent/CN117793608A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5371799A (en) | Stereo headphone sound source localization system | |
| Gardner | Reverberation algorithms | |
| CN102395098B (en) | Method of and device for generating 3D sound | |
| US8126172B2 (en) | Spatial processing stereo system | |
| CN102334348B (en) | Converter and method for converting an audio signal | |
| US11317233B2 (en) | Acoustic program, acoustic device, and acoustic system | |
| EP0735796A2 (en) | Method and apparatus for reproducing three-dimensional virtual space sound | |
| Farina et al. | Ambiophonic principles for the recording and reproduction of surround sound for music | |
| JP2023517720A (en) | Reverb rendering | |
| Su et al. | Inras: Implicit neural representation for audio scenes | |
| JP2012509632A5 (en) | Converter and method for converting audio signals | |
| Genuit et al. | A virtual car: Prediction of sound and vibration in an interactive simulation environment | |
| GB2612173A (en) | Determining a virtual listening environment | |
| Yadav et al. | A system for simulating room acoustical environments for one’s own voice | |
| Krebber et al. | Auditory virtual environments: basics and applications for interactive simulations | |
| Kahana et al. | A multiple microphone recording technique for the generation of virtual acoustic images | |
| CN117793608A (en) | Method for generating space reverberation sense | |
| JP2924502B2 (en) | Sound image localization control device | |
| Sottek et al. | Tools and methods for product sound design of vehicles | |
| Ewert et al. | Computationally-efficient and perceptually-motivated rendering of diffuse reflections in room acoustics simulation | |
| Tamulionis et al. | Listener movement prediction based realistic real-time binaural rendering | |
| Palenda et al. | Setup for choir recordings in virtual churches | |
| Romblom | Diffuse Field Modeling: The Physical and Perceptual Properties of Spatialized Reverberation | |
| JP3468198B2 (en) | Sound field reproduction device | |
| Farina et al. | Numerical model of the sound field inside cars for the creation of virtual audible reconstructions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |