Method for controlling a sound field, related control system, and related computer program

Abstract

emitting sound waves from a set of speakers (12, 14) based on a predetermined signal; measuring the response of the sound waves of the individual or each speaker (12, 14) with a plurality of sensors (16) to obtain initial measurements, each associated with one of the speakers (12, 14) and one of the sensors (16); a definition step for defining a subset of radiating loudspeakers from the set of loudspeakers (12, 14) according to predetermined criteria, the subset including at least one radiating loudspeaker; a determining step of determining a set of transform filters for reproduction of the acoustic signal (50) by a set of speakers (12, 14); 10. A sound field control method comprising a calibration phase comprising:

Description

[Technical Field]

[0001] The present invention relates to a method for controlling a sound field.

[0002] The invention also relates to a sound field control system.The invention also relates to an associated computer program.

[0003] The present invention relates to the area of ​​sound field control. [Background technology]

[0004] Audio reproduction systems are known per se. These systems are intended for professional use, for example by people in mixing studios, or for private use in home theaters (or "home cinemas" in English). Such systems are also used, for example, to control the sound field for large audiences, for example in cinemas or concert halls. These systems integrate, for example, speakers capable of reproducing only mid- or high-frequencies, and speakers capable of reproducing low frequencies as well, and / or speakers dedicated to the reproduction of low frequencies, called "subwoofers" or bass boxes.

[0005] Generally, sound field reproduction takes place in a listening area inside a room whose walls cause numerous reflections and resonances that amplify, attenuate, or eliminate certain frequencies of the sound field in the listening area. Additionally, the listening area may present obstacles that create other reflections and resonances.

[0006] These sound field modifications vary at different locations in the space. To achieve a satisfying auditory experience for users located at different positions within a listening area covered by the sound field, it is necessary to control the sound waves radiated from each speaker. In particular, to obtain a controlled sound field at each position within the listening area, it is necessary to control and modify the radiated sound waves. For example, if not controlled, interference may disrupt the auditory experience.

[0007] Such control depends on numerous parameters, such as the number, position and orientation of speakers in the room in which the sound field is reproduced, the shape of the listening room, the material of the walls, and the location and type of obstacles.

[0008] Also, achieving optimal control for each position in the listening area is often difficult, if not impossible, given the great complexity of the various parameters and factors that affect the control, so known solutions for sound field control still have room for improvement. Summary of the Invention [Problem to be solved by the invention]

[0009] It is an object of the present invention to obtain a sound field control method and an associated control system that makes it possible to obtain a high quality auditory experience, in particular at each position in the listening area, and it is another object of the present invention to obtain a control method that is easy to implement. [Means for solving the problem]

[0010] Towards this end, one goal of the present invention is to: emitting sound waves from a set of speakers in response to a predetermined signal; measuring the response of the sound waves with a plurality of sensors, each for a respective one of the one or more speakers, to obtain an initial measurement, each for one of the one or more speakers and additionally for one of the sensors; - defining a subset of at least one loudspeaker from the set of loudspeakers according to predetermined criteria, said subset of at least one radiating loudspeaker; a processing step including applying a predetermined function to each initial measurement for the at least one radiating loudspeaker to obtain a processed measurement for each initial measurement for the at least one radiating loudspeaker; determining a set of transformation filters for reproduction of the sound signal by a set of loudspeakers, the set of transformation filters being determined in response to at least each initial measurement and in response to a target sound field resulting from each processed measurement; A sound field control method including a calibration phase including:

[0011] According to other advantageous aspects of the invention, the control method comprises one or more of the following features, alone or in any technically possible combination: In the radiating step, a plurality of speakers from the set of speakers radiate sound waves in response to a predetermined signal. The speaker emits a continuous stream of sound waves in response to a predetermined signal. - a control method further comprising a reproduction phase of an acoustic signal comprising the steps of: a filtering step comprising applying, for each loudspeaker, a transformation filter from a set of transformation filters to the acoustic signal in order to obtain a control signal for each loudspeaker; and + a playback step of playing the audio signals by a set of speakers in response to each corresponding control signal. A predetermined function modifies the amplitude of a portion of each initial measurement to obtain each processed measurement. The predetermined function defines at least one time envelope to be applied to each initial measurement. The time envelope selects only the first part of each initial measurement having a predetermined time length, and other parts of each initial measurement are eliminated; and / or The time envelope reduces the amplitude of each initial measurement as a function of time, and / or the time envelope corresponds to the decay time of the acoustic response following each initial measurement, and / or - Temporal envelope increases the amplitude of each initial measurement as a function of time. The predetermined function comprises a frequency filter that modifies the amplitude of each initial measurement depending on the frequency of the initial measurement. The frequency filter is a time-frequency filter that modifies the amplitude of each initial measurement also as a function of time. The calibration phase further comprises a step of integration of the processed measurements to obtain, for each sensor, target measurements that define the target sound field. In a determination step, a set of transformation filters is determined by minimizing a cost function, the parameters of which are at least each of the initial measurements and the target sound field. The cost function determines at least the sum of the errors in the least squares sense for each sensor between the sum of the initial measurements for each loudspeaker filtered by the corresponding transformation filter and the target measurements. The cost function includes at least the following equation:

number

[0012] The invention also has as its object a computer program comprising software instructions which, when executed by a computer, carries out the control method described above.

[0013] The present invention has as its object a control system in which a sound field comprises a set of loudspeakers, a plurality of sensors and at least one control device, the set of loudspeakers being configured to emit sound waves in response to a predetermined signal, and the plurality of sensors being configured to measure the response of the sound waves for the or each loudspeaker individually to obtain initial measurements, each relating to one of the loudspeakers and additionally one sensor; The control system is further configured to: - defining a subset of at least one loudspeaker, said at least one radiating loudspeaker, from among the set of loudspeakers according to predetermined criteria; applying a predetermined function to each initial measurement for the at least one radiating loudspeaker to obtain a processed measurement for each initial measurement for the at least one radiating loudspeaker; determining a set of transformation filters for the reproduction of the sound signal by a set of loudspeakers, the set being determined at least as a function of each initial measurement and as a function of the target sound field resulting from each processed measurement;

[0014] The present invention has as its objectives: - emitting sound waves by a set of speakers in response to a predetermined signal; - measuring the response of the sound waves by the sensors for the or each speaker individually to obtain initial measurements relating to one of the speakers and one sensor each; a definition step for defining a subset of said loudspeakers, each of said loudspeakers being called a radiating loudspeaker, according to predetermined criteria; a processing step of each calculated response corresponding to the acoustic response of an individual radiating loudspeaker of the subset at the location of each of the sensors, the processing step including determining a set of calculated responses of the radiating loudspeakers; - a step of combining said calculated responses to obtain a target response defining a target sound field for each position; - determining a set of transformation filters for the reproduction of sound signals by said set of loudspeakers, said set of transformation filters being determined at least in response to each of said initial measurements and in response to the target sound field obtained from each of said calculated responses; The present invention has a sound field control method including a calibration phase including: [Brief explanation of the drawings]

[0015] These features and advantages of the invention will become apparent from the following description, given by way of non-limiting example only, and made with reference to the accompanying drawings, in which: [Figure 1] FIG. 1 is a schematic diagram of a sound field control system according to the present invention. [Figure 2] FIG. 2 is a flowchart of a control method executed by the control system of FIG. [Figure 3] FIG. 3 is a first example of a speaker arrangement for the control system of FIG. [Figure 4] FIG. 4 is a second example of speaker placement for the control system of FIG. [Figure 5] Figure 5 includes a schematic perspective view of a portion of the control system of Figure 1. Figure 5 also includes two views according to the plane of the schematic perspective view. [Figure 6] FIG. 6 is a schematic diagram of an example of speaker placement in the control system of FIG. [Figure 7] FIG. 7 is a schematic diagram of an example of speaker placement in the control system of FIG. [Figure 8] FIG. 8 is a schematic diagram of an example of speaker placement in the control system of FIG. [Figure 9] FIG. 9 is a schematic diagram of an example of speaker placement in the control system of FIG. [Figure 10] FIG. 10 is a schematic diagram of an example of speaker placement in the control system of FIG. [Figure 11] FIG. 11 is a schematic diagram of an example of speaker placement in the control system of FIG. [Figure 12]FIG. 12 is a diagram of the amplitude of the sound signal at a predetermined position in the sound field according to different frequencies and different times in the absence of filtering of the sound signal by a conversion filter according to the control method of FIG. [Figure 13] FIG. 13 is a diagram of an acoustic signal filtered by a transform filter according to the control method of FIG. 2, similar to the diagram of FIG. DETAILED DESCRIPTION OF THE INVENTION

[0016] Referring to FIG. 1, a sound field control system 10 includes a set of speakers 12 , 14 , a plurality of sensors 16 , and at least one controller 18 .

[0017] The control system 10 is configured to control the sound field in a listening area 19. The listening area 19 is in particular located in a part of a room 20, such as a cinema room, a concert hall or a living room. According to one example, the listening area 19 is identical in size to the room 20.

[0018] The set of speakers includes, for example, at least one radiating speaker 12 and at least one non-radiating speaker 14. According to one example, the radiating speaker 12 and the non-radiating speaker 14 exhibit the same technical characteristics and different functions, in particular for generating a sound field during operation of the control system 10.

[0019] In particular, each loudspeaker 12, 14 of the set of loudspeakers is connected by a wired or wireless connection to a control device 18 for control by the control device 18. For reasons of visibility, only the connection between the control device 18 and the radiating loudspeaker 12 is shown in Figure 1.

[0020] In particular, the radiating loudspeakers 12 are positioned and / or controlled so as to generate a first wavefront that exhibits a predetermined spatial and temporal uniformity across the listening area 19 .

[0021] Preferably, non-radiating speakers 14 do not contribute to this first wavefront, which exhibits a predetermined spatial and temporal uniformity across listening area 19 .

[0022] "Predetermined spatial and temporal uniformity" means, in particular, that interference between the radiating speakers 12 is below a predetermined threshold at each time and at each position in the listening area 19 in a predetermined frequency band.

[0023] The term "first wave front" particularly refers to a set of a position in space and a time at which a wave emitted from a corresponding one or more speakers is first formed. At a predetermined time, the first wave front corresponds to a position in space where the wave first arrives. At a predetermined position in space, the first wave front corresponds to a time at which the wave first arrives at this position. In particular, the first wave front is included in the initial part of a predetermined time length of the sound wave emitted by the corresponding one or more emitting speakers 12.

[0024] The set of speakers 12, 14 is configured to compensate, for example, attenuate, a portion of the sound waves radiated by the radiating speaker 12. In particular, both the radiating speaker 12 and the non-radiating speaker 14 are configured to compensate, for example, attenuate, a portion of the corresponding sound waves.

[0025] 1, control system 10 preferably includes a plurality of radiating speakers 12, e.g., four, and a plurality of non-radiating speakers 14, e.g., four. According to one example, not shown, control system 10 includes a single radiating speaker 12.

[0026] The sensors 16 are, for example, microphones. Each sensor 16 defines a measurement location, preferably in three dimensions. Each sensor 16 is configured to measure sound waves generated by the speakers 12, 14 at the measurement location. The sensors 16 are preferably omnidirectional microphones, whose sensitivity is independent or almost independent of the direction of wave incidence on the microphone.

[0027] The control device 18 is, for example, a computer.

[0028] The control device 18 comprises, for example, a receiving module 21 , a processing module 22 and a transmitting module 24 .

[0029] In the example of FIG. 1, the receiving module 21, the processing module 22 and the transmitting module 24 are each at least partially realized in the form of software or in the form of software blocks that are, for example, stored in the memory 30 of the control device 18 and executable by the processor 32 of the control device 18.

[0030] The memory 30 of the controller 18 can store the receiving software, the processing software, and the transmitting software. The processor 32 of the controller 18 can execute the receiving software, the processing software, and the transmitting software.

[0031] The operation of the control system 10 will now be explained with reference to FIG. 2, which is a flow chart of a method 100 for controlling a sound field according to a first embodiment.

[0032] The control method 100 according to the first embodiment includes in particular a preparation phase 102, a calibration phase 104 and a reproduction phase 106 of the sound signal to be reproduced.

[0033] In the preparation phase 102, the set of speakers 12, 14 and / or the number and / or arrangement of sensors 16 are determined. The preparation phase 102 is executed, for example, by the control device 18 or by a computer different from the control device 18.

[0034] The preparation phase 102 includes, for example, a first preparation decision step 110 and a second preparation decision step 112 .

[0035] In a first preliminary determination step 110, in particular the number and arrangement of the set of loudspeakers 12, 14 is determined so as to obtain a first wavefront exhibiting a predetermined spatial and temporal uniformity radiated by the radiating loudspeakers or from each radiating loudspeaker jointly.

[0036] Several examples of placement of the radiating speakers in the first preliminary decision step 110 are described below.

[0037] According to a first example, in step 110, a single emitting speaker 12 is selected for the control system 10. In this case, the emitting speaker 12, during operation, radiates waves that exhibit a spherical propagation shape on initial approach. The other speakers are non-radiating speakers 14. All speakers 12, 14 jointly compensate for, for example, reflections and resonances of the sound waves radiated by the emitting speaker 12 in the listening area 19 during operation of the control system 10.

[0038] According to a second example, and with reference to FIG. 3, in step 110, two radiating speakers 12 are selected for the control system 10.

[0039] In a second example, the radiating speakers 12 are arranged according to a straight line, for example parallel to the walls of the room 20 .

[0040] Preferably, the listening area 19 corresponds to a sector defined in this example from the midpoint of the line segment joining the two radiating speakers 12, whose angular aperture is inversely proportional to the distance separating the two radiating speakers 12 and / or inversely proportional to the reproduced frequency. This sector thus defines the listening area 19. In the listening area 19 according to this example, a first wavefront, designated by the general reference FR in Figure 3, indicates the direction of propagation indicated by the arrow 40.

[0041] For example, the wavefront emitted by the radiating speaker 12 will contain destructive interference with an attenuation equal to or less than -3 dB as long as the fundamental wavefront of the radiating speaker 12 overlaps with the nearest neighboring wavefront within 0.25 wavelengths.

[0042] In particular, the first wavefront is uniform and / or largely free of interference, and in particular, this first wavefront is furthermore directional and directed towards the listening area 19, while the first wavefront in the direction of the side walls is attenuated by destructive interference, thereby reducing the creation of undesirable room reflections and resonances.

[0043] According to a third example, and with reference to Figure 4, in step 110, the control system 10 selects four radiating speakers 12, i.e., four radiating speakers 12 are placed in the room 20. As in the second example, the radiating speakers 12 are preferably placed according to a straight line, for example parallel to the walls of the room 20.

[0044] For example, the radiating speakers 12 are arranged in a configuration known as a "double bass array" (literally meaning an array of two basses in English), in which case reflections of, for example, four radiating speakers 12 on the side walls of the room contribute to the first wavefront.

[0045] According to a third example, as shown in particular in Figure 4, the radiating loudspeakers 12 are coupled to obtain a first wavefront that substantially forms a plane wave 42. A plane wave is in particular the most developed case of a directional wave.

[0046] In a general example of the invention, the radiating speakers are preferably arranged so that when they are coupled they produce directional waves, and / or in extreme cases unidirectional waves or plane waves.

[0047] Of course, in particular, the sound field develops in three dimensions, and the radiating loudspeakers 12 are arranged in space so as to create a uniform first wavefront in the 3D listening area 19. Figures 3 and 4 are therefore to be interpreted as both top and side views, as shown in Figure 5 for the example of Figure 3.

[0048] For example, referring to Fig. 5, the control system has four radiating speakers 12 arranged in the same plane. According to a first plane A, for example, corresponding to a top view, two of the four radiating speakers 12 are visible, and the other two are superimposed. According to a second plane B, for example, corresponding to a side view, two of the four radiating speakers 12 are also visible. Therefore, the arrangement shown in Figs. 3 and 4 applies to both the arrangement in the first plane A and the arrangement in the second plane B.

[0049] Examples of arrangements of the radiating speakers 12 of the control system 10 are shown with reference to Figures 6 to 11. The radiating speakers 12 according to each of these arrangement examples are arranged and / or controlled so as to be able to generate a first wavefront that exhibits a predetermined spatial and temporal uniformity across the listening area 19, in particular a first wavefront that is uniform in 3D.

[0050] For example, the radiating speakers are arranged in the same plane, such as the vertical plane visualized in FIGS.

[0051] For example, referring to Fig. 6, the control system 10 has six radiating speakers 12, three of each arranged in parallel straight lines. Referring to Fig. 10, the control system 10 has three radiating speakers 12 arranged in straight lines. Referring to Fig. 11, the control system 10 has two radiating speakers 12 arranged in straight lines. Referring to Fig. 7, the control system 10 has three radiating speakers 12 arranged in a triangular shape. Referring to Fig. 8, the control system 10 has five radiating speakers 12 arranged approximately randomly while respecting the maximum distance criterion between adjacent speakers.

[0052] In particular, the maximum distance between two adjacent radiating speakers is such that the waves radiated from each of the two adjacent radiating speakers overlap within 0.25 wavelengths in the angular range of the listening area. This distance can be estimated, for example, by the following approximate calculation:

number

[0053] For example, for a listening area 19 with an angular extent of 60°, and for a shortest wavelength of 3.44 m, the maximum distance between the two speakers is 1.72 m, corresponding to a maximum frequency of 100 Hz.

[0054] Referring to FIG. 9, the control system 10 has four radiating speakers 12 arranged according to a rectangle.

[0055] In a second preliminary determination step 112, the number and placement of the sensors 16 are determined. For example, the measurement position of each sensor 16 is determined to obtain the placement of the sensors 16.

[0056] In particular, the set of sensors 16 are arranged such that the measurement positions of the sensors 16 allow an unambiguous measurement of the sound field to be obtained.

[0057] By "unique" we mean that the signals from the sensors 16 represent at least the amplitude of the sound field at each location in the listening area 19, not just at the sensor locations 16, up to a predetermined maximum frequency, called the spatial aliasing frequency. A continuous sound field in space can be accurately represented by a finite number of sensors without loss of information. In particular, adding an additional sensor 16 does not provide additional information about the sound field, such as amplitude values ​​for a predetermined frequency and a predetermined location in the listening area 19 at a predetermined time, compared to the information already contained in the signal from the sensor 16 in the absence of the additional sensor.

[0058] For example, the sensors 16 are arranged according to an irregular mesh in three dimensions, where "irregular mesh" means in particular that in three dimensions each distance may be different from every other distance between the sensors 16 and / or in particular that the distance between at least two adjacent sensors 16 exhibits a different distance compared to the distance between any other two adjacent sensors 16.

[0059] Preferably, the distance between two selected adjacent sensors 16 is equal to or less than half the shortest wavelength controlled in the sound field to satisfy the conditions of the Nyquist-Shannon theorem. In other words, for example, the placement of the sensors 16 is determined by arranging each sensor 16 so that the maximum distance between the sensors 16 and their nearest neighbors is strictly shorter than half the shortest wavelength in the sound field.

[0060] For example, for a maximum frequency of 100 Hz, the maximum distance between two adjacent sensors 16 is 1.72 meters.

[0061] Preferably, the placement of sensors 16 is non-planar. By "non-planar," it is meant that there is no plane in three-dimensional space that contains all of sensors 16. For example, at least one sensor 16 is positioned outside of such a plane.

[0062] The number of sensors 16 selected in step 112 depends, among other things, on the maximum frequency of the sound field to be controlled and the three-dimensional dimensions of the listening area 19. For example, when the distance between two adjacent selected sensors 16 is less than half the shortest wavelength to be controlled in the sound field, the three-dimensional dimensions of the listening area 19 will determine the number of sensors 16.

[0063] According to one example, the sensors 16 are arranged according to a tetrahedral mesh in 3D, or according to any other regular polyhedron, or according to a 3D mesh formed from multiple rows and layers of staggered sensors. Preferably, the mesh is selected irregularly. Alternatively, a regular mesh can be made irregular by applying a random displacement to the position of each sensor.

[0064] For example, the sensor 16 is positioned to satisfy the conditions of the Nyquist-Shannon theorem.

[0065] In a calibration phase 104 , the sound field is calibrated in response to measurements obtained, among other things, from the sensors 16 .

[0066] Preferably, the calibration phase 104 is performed at least once before the regeneration phase 106 is performed.

[0067] The calibration phase 104 includes, for example, a radiating step 120 , a measuring step 122 , a defining step 123 , a processing step 124 , an integrating step 126 , and a determining step 128 .

[0068] In a emitting step 120, the or each speaker 12, 14 receives a predetermined signal and consequently emits sound waves in response to the predetermined signal.

[0069] Preferably, the speakers 12, 14 receive a predetermined signal and emit sound waves sequentially, so that one of the speakers 12, 14 emits a sound wave, followed by another speaker 12, 14, and so on, until all of the speakers 12, 14 in the control system 10 are measured.

[0070] The predetermined signal may be, for example, a spectrally dense and / or deterministic excitation signal, e.g., the predetermined signal may have multiple frequencies in a predetermined frequency range, or may comprise an excitation signal whose amplitude and / or frequency is modified over time.

[0071] For example, the predetermined signal may include, and preferably consist of, frequencies equal to or less than a predetermined maximum threshold, such as 100 Hz or 200 Hz.

[0072] In a measuring step 122, each sensor 16 measures the sound wave response for the or each speaker 12, 14 individually to obtain an initial measurement. Each initial measurement is associated with one of the one or more speakers 12, 14 and also with one of the sensors 16. Thus, in particular, each initial measurement corresponds to the sound wave response measured by one of the sensors 16 that is considered to be emitted from only one of the speakers 12, 14 or emitted individually.

[0073] In a definition step 123, the control device 18 defines a subset of the set of loudspeakers 12, 14, comprising at least one loudspeaker that is an emitting loudspeaker 12. The control device 18 defines the or each emitting loudspeaker 12 according to predetermined criteria, in particular that the emitting loudspeaker or loudspeakers 12 are capable of generating a first wavefront that exhibits a predetermined spatial and temporal uniformity over the listening area 19.

[0074] According to one example, the emitting loudspeaker or a subset comprising each emitting loudspeaker 12 forms the parameters received by the control device 18 .

[0075] Preferably, the predetermined criteria define that the radiating loudspeaker 12 is a loudspeaker positioned to generate a first wavefront in the listening area 19 that exhibits interference below a predetermined threshold between the fundamental wavefronts radiated individually by the radiating loudspeakers 12 for a predetermined frequency band at each time and each position in the listening area 19.

[0076] In particular, the radiating speakers 12 defined according to predetermined criteria are arranged according to the arrangement examples of FIGS.

[0077] In processing step 124, the control device 18, and in particular the processing module 22, applies a predetermined function to each initial measurement associated with each radiating loudspeaker 12 in order to obtain a processed measurement for each of these initial measurements. In particular, the control device applies the same predetermined function to each initial measurement. In particular, the control device 18 applies the predetermined function to each initial measurement associated with one radiating loudspeaker 12 from among the radiating loudspeakers 12 in order to obtain a processed measurement for each initial measurement associated with the corresponding radiating loudspeaker 12.

[0078] Preferably, the predetermined function modifies the amplitude of a portion of each initial measurement to obtain each processed measurement.

[0079] According to one example, the predetermined function comprises a frequency filter that modifies the amplitude of each initial measurement depending on frequency, for example by attenuating certain frequency bands and amplifying other frequency bands.

[0080] According to one example, the frequency filter is a time-frequency filter that modifies the amplitude of each initial measurement also as a function of time.

[0081] For example, the predetermined function defines at least one time envelope that is applied to each initial measurement.

[0082] The time envelope may, for example, be selected by selecting only the first part of each initial measurement having a predetermined time length, while other parts of each initial measurement are removed. This corresponds in particular to truncating the initial measurements. For example, only the first wavefront of the sound field as measured according to each initial measurement is extracted, and in particular subsequent sound waves are removed. This makes it possible, for example, to remove or at least reduce the response of a room 20 included in the initial measurement and present in the listening area 19.

[0083] According to one example, the time envelope decreases or increases the amplitude of each initial measurement over time, which can, in particular, reduce, but not completely eliminate, the duration of the response of room 20, or increase the duration of this response.

[0084] According to one example, the time envelope may contain values ​​greater than 1 for certain times, so that each initial measurement can be amplified at those times.

[0085] According to one example, the temporal envelope comprises a value greater than 1 and increasing with time, which allows lengthening the response of the room 20, in particular if, for example, the response is too short or the room effect is amplified.

[0086] According to one example, the time envelope corresponds to the decay (from the English term "time decay", which literally means "decay time") of the sonic response according to the initial measurement. The decay depends in particular on the frequency of the sonic response according to the initial measurement. For example, according to this example, the time envelope reduces the decay for certain frequencies and amplifies the decay for other frequencies. According to another example, the time envelope adjusts the decay to obtain a uniform decay according to frequency.

[0087] In an integration step 126, the controller 18, and in particular the processing module 22, integrates the processed measurements to obtain, for each sensor 16, a target measurement that defines the target sound field at this location.

[0088] In particular, the control device 18 individually combines the processed measurements of the measurement positions of the corresponding sensors 16 to obtain a target measurement at this measurement position. For example, for a given sensor 16, the control device 18 combines the processed measurement obtained from the initial measurement of the first emitting speaker 12 with the processed measurement obtained from the initial measurement of the second emitting speaker 12, and with the processed measurements obtained from the initial measurements of all emitting speakers 12. Thus, in particular, each target measurement is associated with a measurement position of the corresponding sensor 16, for example as determined in step 112.

[0089] When the measurement locations of the sensors 16 are positioned to obtain unambiguous measurements of the sound field, the target measurements at these locations substantially represent the target sound field in the listening area 19 .

[0090] According to one example, combining the processed measurements includes summing the processed measurements to obtain each target measurement.

[0091] In this case, for example, the target sound field corresponds to a first wavefront that is uniform and / or interference-free and / or directed towards the listening area 19, especially when the initial measurement has been processed to extract only the first wavefront, for example by truncation. Preferably, in this case, the wavefronts that are directed outside the listening area 19, for example towards the walls of the room 20, are attenuated by destructive interference so that in particular resonances and reflections of the room 20 are hardly or not excited at all.

[0092] According to another example, or in addition, the integration of the processed measurements comprises applying, for example, a gain and / or a certain delay before the addition of the processed measurements, which makes it possible, in particular, on the one hand, to adjust the width and / or shape of the first wavefront and in particular to control the excitation conditions of the room, and, on the other hand, to compensate for inaccurate positioning of the radiating loudspeakers.

[0093] According to another example, or in addition, combining the processed measurements includes applying a filter that modifies the amplitude, phase, and / or delay of the sound waves of the individual processed measurements, for example, the filter modifies the amplitude, phase, and / or delay individually for the processed measurements of each radiating loudspeaker 12. The filter may, for example, be configured using a beamforming technique (literally meaning "beam formation"), which allows, in particular, to control the direction and / or width of the first wavefront generated by the combined action of the radiating loudspeakers 12.

[0094] The integration step 126 is an optional step.

[0095] Preferably, the integration step 126 is performed in the presence of multiple radiating speakers 12 .

[0096] For example, the integration step 126 is performed when the control system 10 has multiple radiating speakers 12 .

[0097] In a decision step 128, the control unit 18 determines a set of transformation filters for transforming the acoustic signal 50 by the or each speaker 12, 14 in response to at least each initial measurement and in response to the target sound field resulting from each processed measurement.

[0098] For example, the control device 18 determines the transformation filter by minimizing a cost function, the parameters of which are, inter alia, at least each initial measurement considered separately for each measurement position and each loudspeaker 12, 14, and the target sound field at each measurement position.

[0099] For example, the cost function determines at least the sum of the errors in the least squares sense for each sensor 16 between the sum of the initial measurements filtered by the corresponding transformation filter and the target measurements.

[0100] For example, the cost function is defined as follows:

number

[0101] For example, let p be the index of one of the measurement positions P of the sensor 16, and let s be the index of one of the speakers S12, 14.

[0102] In particular, the vector g is defined as follows:

number

[0103] Each element g s is from t=1 to t=T g A dimension T containing the impulse response of the transform filter for speaker s for each time g is a vector of

number

[0104] The overall convolution matrix H is, for example, a multi-channel convolution matrix for the measurement position P of the speakers S12, S14 and the sensor 16, and is defined for each block as follows, for example.

number

[0105] Convolution matrix h ps For example, is defined as follows:

number

[0106] For example, the convolution matrix is ​​a Toeplitz matrix.

[0107] For example, when two speakers from the set of speakers are selected as speakers 12, 14 and the control system 10 has two sensors 16, the overall convolution matrix is ​​defined block by block as follows: H=[h 11 h 12 ;h 21 h22 ] Here, ";" means vertical concatenation.

[0108] The target response d of the target sound field is defined, for example, as follows:

number

[0109] Each element d p In particular, from t=1 to t=T g The dimension T contains the impulse response of the target sound field at position p for each time up to g is a vector of

number

[0110] According to another example, or in addition, the cost function is defined as follows:

number

[0111] According to another example, or in addition, the cost function may include N regularization terms Reg, each of which depends at least on g, and preferably also on other parameters Γ, which may include one or more regularization matrices therein. n (g) includes the set.

number

[0112] Following the example, the vectors may be defined as follows:

number

[0113] Or it is defined as follows:

number

number

[0114] The reconstruction phase 106 of the acoustic signal 50 includes, for example, a filtering step 140 and a reconstruction step 142 .

[0115] In a filtering step 140, the control unit 18, and in particular the processing module 22, applies a conversion filter to the acoustic signal 50 for each speaker 12,14 to obtain a control signal for that speaker 12,14.

[0116] In a playback step 142, the set of speakers 12, 14 plays the audio signal 50 in response to their corresponding control signals.

[0117] When the reproduction phase 106 involves the reproduction of an acoustic signal 50 that includes a low frequency range, for example below 100 Hz, according to one example, the reproduction phase 106 includes a pre-processing step that involves extracting portions of the signal that have a frequency range below 100 Hz, in which case, for example, only these portions are applied to a transform filter in a filtering step 140.

[0118] Preferably, each speaker 12, 14, and in particular each radiating speaker 12, radiates sound waves comprised of frequencies equal to or less than a predetermined maximum threshold, for example 100 Hz or 200 Hz. This is preferably associated with each phase 102, 104, 106 of the method 100. In particular, each speaker 12, 14 is configured to radiate only sound waves with frequencies below the maximum threshold.

[0119] A second embodiment of the control method 100 is described below. The control method 100 according to the second embodiment includes at least some of the features of the first embodiment, except for the differences described below.

[0120] The calibration phase 104 of the control method 100 according to the second embodiment includes, for example, a emitting step 120 , a measuring step 122 , a defining step 123 , a processing step 124 , an integrating step 126 , and a determining step 128 .

[0121] According to one example, the defining step 123 comprises defining a subset comprising a plurality of radiating speakers 12 from the set of speakers 12, 14 according to predetermined criteria.

[0122] The processing step 124 according to the second embodiment includes, for example, determining a set of calculated responses of the radiating loudspeaker 12 .

[0123] Each calculated response corresponds in particular to the acoustic response of each radiating loudspeaker 12 of the subset at the position of each sensor 16, in particular at positions called measurement positions in the first embodiment. The form of each calculated response is in particular a composite response.

[0124] For example, each calculated response corresponds to an impulse located on a first wavefront radiated by the corresponding radiating loudspeaker 12. The impulse of each calculated response depends, among other things, on the propagation time between the position of the corresponding radiating loudspeaker 12 and the measurement position. In particular, the set of calculated responses represents an idealized wavefront of an omnidirectional sound source.

[0125] In an integration step 126, the controller 18 integrates the calculated responses for each sensor 16, and in particular for each measurement location, to obtain a target response that defines the target sound field at this location. For example, the controller 18 integrates the calculated responses in the same way as it integrates the processed measurements according to the first embodiment. In this case, the controller 18 obtains a target response instead of the target measurements defined in the first embodiment.

[0126] For example, combining the calculated responses may involve summing the calculated responses at each measurement location separately to obtain a respective target response at this location.

[0127] In a determining step 128, the controller 18 determines a set of transformation filters for the reproduction of the acoustic signal 50 by the set of speakers 12, 14. According to a second embodiment, the set of transformation filters is determined in response to at least the respective initial measurements and in response to the target sound field resulting from the respective calculated responses, in particular in this case the target sound field being defined by the target response.

[0128] In particular, the control method 100 according to the second embodiment includes determining a calculated response instead of the processed measurement according to the first embodiment. The control method 100 according to the second embodiment includes, for example, determining a target response instead of a target measurement.

[0129] In particular, the control method 100 according to the second embodiment includes some or all of the features of the method according to the first embodiment, including replacing the target measurement with a target response.

[0130] According to one example, the control method 100 according to the second embodiment includes the application of a predetermined function to obtain a calculated response.

[0131] The preparation phase 102 and / or the regeneration phase 106 according to the second embodiment are preferably identical to those of the first embodiment, in particular with the application of the differences described above for the second embodiment.

[0132] 12 and 13 are diagrams including the amplitude A in dB according to different frequencies F in Hz of an acoustic signal at a given position in the sound field, shown over a time t in ms. FIG. 13 represents the amplitude after application of the conversion filter, while FIG. 12 represents the amplitude without application of the conversion filter. It can be seen that the amplitude A is attenuated when the conversion filter is applied from t=100 ms, thus corresponding to an example of truncation of the sound wave, which in particular makes it possible to reduce or eliminate reflections and resonances of the sound wave. When applying the conversion filter, the decay duration of the listening room is significantly reduced. In particular, by using the control method, an attenuation of the amplitude A and therefore a reduction in the decay is obtained at each position in the listening area 19.

[0133] It will be appreciated that the control method 100 and control system 10 exhibit numerous advantages.

[0134] The use and processing of the initial measurements and possibly their integration and / or the determination of a set of calculated responses to obtain a transformation filter and their integration allows obtaining a high-quality acoustic experience, in particular over the listening area 19. According to preferred embodiments, this is valid not only at the measurement location of the sensor 16 but also outside it, due to the unambiguous representation of the sound field. The measurements also allow the method to easily adapt the control of the loudspeakers 12, 14 individually to the listening area 19.

[0135] In particular, the control method 100 is very robust to potential measurement noise. Indeed, while measurements at low frequencies, e.g., equal to or lower than 100 Hz, are very noisy, the same noise is present in the target sound field by construction, and therefore does not directly affect the transform filter. This avoids resorting to high levels of regularization, particularly in the useful bandwidth, and thus simplifies the method.

[0136] Also, according to an example, control method 100 does not require the determination of propagation delays, which are inaccurate especially at low frequencies and significantly impair performance in the case of idealized responses.

[0137] Furthermore, the integration step 126 allows for a target sound field that is easily achievable by integrating the measurements or responses from the radiating loudspeakers 12. Therefore, it is a result that can be achieved by at least the radiating loudspeakers 12 in terms of construction. In the decision step 128, transformation filters are calculated for a target close to the initial measurements, resulting in simple filters that minimize the cost function. These filters are very simple, especially in the initial part, and do not contain audible artifacts such as significant amplification and long response times. This results in a significant improvement in sound quality. In particular, the integration step 126 eliminates the need to use an overall response to define the target sound field, which would deviate from the initial measurements, resulting in a large error in the cost function and the creation of more complex filters. In particular, the integration of the responses from the radiating loudspeakers 12 is applied to the filtered initial measurements, i.e., the processed measurements, and the composite response, i.e., the response calculated for each radiating loudspeaker 12.

[0138] The control method 100 makes it possible to significantly reduce the variations in the amplitude of the sound field obtained during the reproduction of the acoustic signal 50 compared to the reproduction of the acoustic signal 50 without applying the conversion filter, in particular over a predetermined frequency range, for example between 20 Hz and 100 Hz. In particular, applying the conversion filter obtained in the calibration phase 104 makes it possible to reduce the variations in the amplitude of the sound waves at each position in the listening area 19.

[0139] The control method 100 can attenuate reflections and resonances during the reproduction of the acoustic signal 50 and reduce the decay duration compared to the reproduction of the acoustic signal 50 without applying the conversion filter, particularly over a predetermined frequency range, for example between 20 Hz and 100 Hz. In particular, applying the conversion filter obtained in the calibration phase 104 can reduce the variation in the decay duration of the listening room at each position in the listening area 19.

Claims

1. a step (120) of emitting sound waves by each of the set of speakers (12, 14) based on a predetermined signal; a step (122) of measuring the response of the sound waves by a plurality of said sensors (16) for each of said speakers (12, 14) to obtain initial measurements, each relating to one of said speakers (12, 14) and one sensor (16); a definition step (123) of defining, according to predetermined criteria, a subset of radiating loudspeakers from the set of loudspeakers (12, 14), said subset comprising at least one radiating loudspeaker; a processing step (124) comprising applying a predetermined function to each of the initial measurements for at least one of the radiating loudspeakers (12) to obtain a processed measurement for each of the initial measurements for at least one of the radiating loudspeakers (12); a determining step (128) of determining a set of transformation filters for reproduction of the acoustic signal (50) by the set of speakers (12, 14), the transformation filters being determined in response to at least each initial measurement and in response to a target sound field resulting from each processed measurement; A sound field control method (100) comprising a calibration phase (104) comprising:

2. 2. The control method (100) of claim 1, wherein in the emitting step (120), a plurality of the speakers (12, 14) from the set of speakers emit sound waves in response to the predetermined signal, and preferably the speakers (12, 14) emit sound waves sequentially in response to the predetermined signal.

3. a filtering step (140) comprising applying, for each loudspeaker (12, 14), a transformation filter from the set of transformation filters to the acoustic signal (50) to obtain a control signal for each loudspeaker (12, 14); a playing step (142) of playing said audio signals (50) through said set of speakers (12, 14) in response to each corresponding said control signal; 3. The control method (100) of claim 1 or 2, further comprising a reproduction phase (106) of the acoustic signal (50) comprising:

4. A control method (100) according to any one of claims 1 to 3, wherein said predetermined function modifies the amplitude of a portion of said each initial measurement to obtain said each processed measurement.

5. the predetermined function is applied to each of the initial measurements. a time envelope in which only a first portion of each of the initial measurements having a predetermined time length is selected, and other portions of each of the initial measurements are removed; and / or a time envelope that reduces the amplitude of each of the initial measurements as a function of time; and / or a time envelope corresponding to the decay time of the acoustic response according to each of the initial measurements; and / or a time envelope that increases the amplitude of each of said initial measurements as a function of time; 5. The control method (100) of claim 4, further comprising defining at least one time envelope of:

6. the predetermined function includes a frequency filter that modifies the amplitude of each of the initial measurements according to the frequency of the initial measurement; Control method (100) according to claim 4 or 5, wherein the frequency filter is preferably a time-frequency filter that also modifies the amplitude of each initial measurement as a function of time.

7. 7. The control method (100) of claim 1, wherein the calibration phase (104) further comprises a step (126) of integrating the processed measurements to obtain, for each sensor (16), target measurements that define the target sound field.

8. 8. The control method (100) of claim 7, wherein in the determining step (128), the set of transformation filters is determined by minimizing a cost function, the parameters of which are at least the initial measurements and the target sound field.

9. the cost function determines at least the sum of errors in the least squares sense for each sensor (16) between the sum of the initial measurements for each speaker filtered by the corresponding transform filter and the target measurements; Preferably, the cost function includes at least the following equation: [Equation 1] where g is a vector containing the transform filters of the set of transform filters for each loudspeaker (12, 14), H is a global convolution matrix including each of the initial measurements of each of the speakers (12, 14) measured by each of the sensors (16); 9. The control method (100) of claim 8, wherein d is a vector containing each target measurement of the target sound field at each location of each sensor (16).

10. the set of sensors (16) being positioned to obtain an unambiguous measurement of the sound field; Preferably, each sensor (16) is positioned such that the maximum distance to its nearest neighbor is strictly less than half the shortest wavelength of the sound field.

11. 11. A control method (100) according to any one of claims 1 to 10, wherein the subset comprises a plurality of radiating speakers (12) arranged in a predetermined listening area (19) to generate, for a predetermined frequency band at each time and each position in the listening area (19), first wave-fronts that exhibit interference below a predetermined threshold between fundamental wave-fronts radiated individually by the radiating speakers (12).

12. emitting (120) sound waves by the set of speakers (12, 14) in response to a predetermined signal; a step (122) of measuring the response of the sound waves by the sensors (16) for the or each speaker (12, 14) individually to obtain initial measurements, each relating to one of the speakers (12, 14) and one sensor (16); a definition step (123) of a subset of the loudspeakers (12, 14), each loudspeaker of which is called a radiating loudspeaker (12), according to predetermined criteria; a processing step (124) of each calculated response corresponding to the acoustic response of an individual radiating speaker (12) of the subset at the location of an individual sensor (16), including determining a set of calculated responses of the radiating speakers (12); a step (126) of aggregating said calculated responses to obtain target responses defining a target sound field for each position; a determining step (128) of determining a set of transformation filters for reproduction of an acoustic signal (50) by the set of speakers (12, 14), the transformation filters being determined in response to at least each of the initial measurements and in response to the target sound field obtained from each of the calculated responses; A sound field control method (100) comprising a calibration phase (104) comprising:

13. The sound field is controlled within a listening area (19) by a control system (10) including a set of speakers (12, 14), a plurality of sensors (16), and at least one control device (18), the set of speakers including at least one radiating speaker (12) and at least one non-radiating speaker (14), the at least one radiating speaker (12) being positioned and controlled to generate a first wavefront exhibiting spatial and temporal uniformity across the listening area (19), the at least one non-radiating speaker (14) not contributing to this first wavefront, and all of the 13. A control method (100) according to any one of claims 1 to 12, wherein the loudspeakers (12, 14) compensate for reflections and resonances of sound waves emitted from at least one of the radiating loudspeakers (12) within an audience area (19), the method comprising a preparation phase (102) in addition to the calibration phase (104), the preparation phase (102) comprising determining the number and / or arrangement of the loudspeakers of the set of loudspeakers (12, 14) and / or the sensors of the plurality of sensors (16) in order to obtain the first wavefront exhibiting a predetermined spatial and temporal uniformity emitted from the at least one radiating loudspeaker.

14. 14. The control method (100) of claim 13, further comprising, in addition to the preparation phase (102) and the calibration phase (104), a reproduction phase (106) of the acoustic signal (50) to be reproduced, the reproduction phase (106) comprising a filtering step (140) and a reproduction step (142), wherein in the filtering step (140) the control device (18) applies, for each of the speakers (12, 14), a conversion filter for the speaker (12, 14) to the acoustic signal (50) to obtain a control signal for the speaker (12, 14) to be reproduced, and in the reproduction step (142) all of the speakers (12, 14) reproduce the acoustic signal (50) to be reproduced in accordance with the corresponding control signal.

15. 15. A control method according to claim 13 or 14, wherein the radiating speakers (12) are arranged such that the maximum distance between two adjacent radiating speakers is such that the waves radiated from the two adjacent radiating speakers overlap within a length of 0.25 times the wavelength in an angular range of the listening area (19).

16. The radiating speaker (12) This is a configuration called "double bass arrangement," In the same plane, According to the rectangle, or Following a straight line, preferably parallel to the wall of the room (20) in which the listening area (19) is located, The control method according to any one of claims 13 to 15, wherein

17. 17. A control method according to any one of claims 13 to 16, wherein the integration to define the target sound field comprises the application of filters to modify the amplitude, phase and / or delay of sound waves of either the processed measurements or the calculated responses of each of the radiating speakers (12), said filters preferably being constituted by beamforming techniques capable of steering and / or controlling the width of a first wavefront generated by the integrated operation of the radiating speakers (12).

18. A computer program comprising software instructions which, when executed by a computer connected to a set of speakers and a plurality of sensors, performs the control method (100) according to any of claims 1 to 17.

19. A control system (10) for a sound field, comprising a set of speakers (12, 14), a plurality of sensors (16) and at least one control device (18), and adapted to carry out a control method according to any of claims 1 to 17.

Images