Chamber for diffusing sound by reflection, method for generating spatialised sound and associated home cinema apparatus

Abstract

The invention relates to a chamber (10) for diffusing sound by reflection, intended to diffuse sound waves in a space comprising at least one planar reflection surface, forming a reflection plane; the chamber for diffusing sound by reflection comprising a frame (32) which comprises a mounting surface for loudspeakers (36); an array of at least two juxtaposed loudspeakers (36) mounted on the mounting surface (34) of the frame, the loudspeakers (36) being arranged at an angle of inclination (P); and a control system (42) configured to send a specific electrical signal to each loudspeaker (36) and to control the amplitude and phase of the electrical signals according to a spatial filtering method. The chamber (20) further comprises a waveguide (38) having at least one acoustic wall (40), the waveguide being arranged in front of the loudspeakers (36).

Description

[0001] REFLEXIVE SOUND DIFFUSION SPEAKER, SPATIALIZED SOUND GENERATION METHOD AND ASSOCIATED HOME CINEMA INSTALLATION

[0002] FIELD OF INVENTION

[0003] The invention relates to a speaker, and more particularly to a sound-reflecting speaker, that is, a speaker designed to project sound waves to a listening point using at least one reflection off a reflective surface, such as a ceiling or wall. The invention also relates to a method for generating spatialized sound and a home theater system. Furthermore, the invention can also be applied to multichannel music playback systems.

[0004] The invention is particularly suitable for multichannel sound systems used, for example, in home theater installations

[0005] STATE OF THE ART

[0006] In the entertainment sector, sound systems paired with large screens allow for home movie viewing with a quality that rivals that of movie theaters. Film soundtracks can now be recorded and played back with immersive sound, also known as "surround sound." Dolby Atmos® technology, developed by Dolby Laboratories®, is arguably the most renowned for cinematic reproduction in digital home theaters. In addition to the horizontal Dolby Atmos® 5.1 or 7.1 format, which provides a perception of sound from the sides and behind the viewer, vertical spatialization is also possible with Dolby Atmos® 5.1.4 or 7.1.4 formats, using four speakers positioned at a height. This technology enables faithful sound reproduction in a three-dimensional space.

[0007] However, while audio enthusiasts can install complex speaker setups in dedicated rooms, the majority of consumers watch movies in rooms that also serve as living spaces, such as a living room or bedroom, or in spaces where they don't have the option of integrating multiple speakers. They therefore look for compact or even centralized systems.

[0008] To meet this need, soundbars have gained popularity, offering better sound quality than most speakers integrated into flat screens, and being able to incorporate technologies that even allow for the reproduction of virtual spatialized sound through special processing and specific speaker arrangements.

[0009] These technologies allow for the simulation of a speaker's placement on a wall or ceiling, creating a "phantom speaker." Thus, instead of receiving only sound waves directly from a speaker facing the audience, the audience can also receive one or more sound waves reflected off a surface by a speaker located away from that reflective surface. The term "phantom speakers" therefore refers to a sound wave perceived from a speaker whose position is simulated by at least one reflection off a reflective surface, typically a wall or ceiling.

[0010] As illustrated in Figure 1 of the prior art, a reflected sound speaker 100 is placed on the floor of a home theater. The room comprises a floor 12, a ceiling 14, and side walls (not shown). In this example, the speaker 100 emits sound that is reflected by the ceiling 14. More precisely, the speaker 100 comprises a chassis 16 on which a loudspeaker 18 is mounted, configured to emit sound towards the ceiling and create a phantom sound that is perceived by a listener from a listening position Pe. In this case, the ceiling 14 acts as a reflecting surface, generally flat. To create a phantom sound from the listening position Pe, the loudspeaker 18 is preferably positioned on a mounting surface with an angle of inclination α between the axis of revolution Ar of the loudspeakers and the reflecting surface.The mounting surface can be flat or inclined so that the angle of inclination a is classically measured with respect to the reflection surface, more precisely with respect to a plane parallel Ppp to the ceiling 14 in the example of figure 1.

[0011] A crosstalk problem exists between the origin of the sound source and the phantom speaker, for example, in thin soundbars. For the purposes of this invention, crosstalk corresponds to the ratio between the reflected background amplitude Sr at the listener's position and the background amplitude Sd at the same position originating directly from the speaker, as illustrated in Figure 1. Indeed, from the same loudspeaker, the propagation time of reflected sound waves is greater than the propagation time of directly transmitted sound waves. The presence of directly transmitted sound waves Sd then disrupts the listening quality of the phantom speakers and limits the spatialization of the sound. Furthermore, if the source's directivity is insufficient, the sound pressure emitted towards Sd and Sr is similar.In this case, the difference in distance traveled between the two waves is not significant enough, so the background sound attenuation SR is not sufficient for the latter to be negligible compared to the sound wave Sd.

[0012] To limit crosstalk, it is therefore necessary to reduce the direct transmission of sound from loudspeakers intended to generate Sr sounds by reflection.

[0013] To address this crosstalk problem, several distinct techniques exist.

[0014] As illustrated in Figure 1 of the prior art, document FR 3105692 proposes combining a reflected sound-diffusion loudspeaker 18 with a waveguide to control directivity and limit the directly transmitted background sound amplitude Sd. However, this solution is limited in effectiveness to certain frequency ranges, particularly low and mid frequencies, due to compactness constraints.

[0015] Furthermore, for a loudspeaker to be directional, its radiating surface must be large relative to the wavelength. A waveguide increases this radiating surface, but even at low frequencies, it is particularly difficult to implement a waveguide in a compact enclosure. Moreover, waveguides typically exhibit variations in directivity, and a waveguide with constant directivity would be difficult to integrate into a compact enclosure due to its height, which is typically greater than 25 cm.

[0016] Another completely different solution proposes to use several speakers and control them in amplitude and phase according to a spatial filtering method to improve the directivity of the sound beam coming from the speakers.

[0017] The frequency range in which directivity control operates depends on the size of the array and the distance between the loudspeakers. To control directivity at low frequencies, a large number of loudspeakers or large loudspeakers are necessary to increase the overall length of the array. This solution is therefore not compatible with a compact system. These systems are also limited at high frequencies because the distance between the loudspeakers must remain small compared to the wavelength. Thus, as the frequency increases, secondary sound beams, or side lobes, appear, which can be directed towards the listener, reducing the level of crosstalk.

[0018] Another solution proposes creating interfering waves with the directly transmitted Sd sound waves in order to cancel or reduce them. For example, document WO 2020 / 102183 describes a loudspeaker system for reproducing a height sound image. To limit crosstalk, the system also includes a loudspeaker arranged at floor level. This loudspeaker is configured to emit a cancellation sound wave, in opposite phase to that of the reflecting loudspeaker, so as to suppress the directly transmitted Sd sound wave, i.e., without reflection, from the reflecting loudspeaker.

[0019] This solution is ineffective due to the excessive distance between the sources, and complex because it requires knowledge of the environment in order to obtain effective digital filtering and limit crosstalk.

[0020] Similarly, in document WO 2020 / 181288, a soundbar includes ceiling-facing spatial speakers and front speakers that emit sound directly towards a listener. Some of the front speakers are used to create a waveform that suppresses the direct component of the sound wave emitted by the spatial speakers. The spatial speakers are phase-controlled to improve the directivity of the resulting sound signal and limit the formation of a direct sound wave towards the listener.

[0021] This solution once again offers a complex filtering technique, and uses additional speakers to limit crosstalk of the signal located in the ceiling.

[0022] Whatever solution is used and in view of the low effectiveness of known solutions, it should be noted that, to limit crosstalk, commercial speakers use a tilt of the speakers with an angle of inclination between 70 and 90° with respect to the reflection surface.

[0023] However, this limited angle of inclination means that the reflective sound diffuser must be placed less than two meters from the audience for the sound waves reflected off a ceiling or wall to be perceived correctly. Therefore, there is a need to extend the reflective sound diffusion distance, for example, to accommodate rooms where the listening position (Pe) is more than two meters from the reflective sound diffuser.

[0024] The technical problem of the invention is therefore to obtain a solution to generate one or more phantom speakers covering a wide range of frequencies, usable for compact systems, with limited crosstalk and offering better flexibility in listener positioning.

[0025] DESCRIPTION OF THE INVENTION

[0026] The invention proposes to address this technical problem by using the cooperation of a waveguide and an array of at least two juxtaposed loudspeakers which can be controlled according to a spatial filtering method to generate a beam of sound waves, also called "beam-forming" in the Anglo-Saxon literature.

[0027] With these elements, the speakers are mounted on the chassis with a high angle of inclination, between 30° and 70° between the axis of revolution of the speakers and the reflecting surface, that is to say with a different angle from the angle seen on commercial products.

[0028] A major problem encountered when projecting a beam of reflected sound waves is that it must be neither too wide, to limit crosstalk, nor too directional, to avoid limiting the listener's positioning. The invention required extensive research to evaluate the factors that could improve or limit the directivity of a reflected sound wave beam over a wide frequency range.

[0029] It is possible to obtain a beam of sound waves with improved directivity by using a spatial filtering method with a loudspeaker array or a waveguide. Contrary to expectations, it has been observed that combining a waveguide and a spatial filtering method enhances their respective performance and goes beyond simply combining these two technologies.

[0030] Indeed, even at low frequencies, a waveguide with small dimensions is often insufficient to achieve good directivity. The presence of a waveguide combined with a spatial filtering technique improves the directivity of a sound wave beam at low frequencies. It has been found that, even at a specific tilt angle, the generation of a reflected sound wave beam is still limited by the presence of high-frequency side lobes emitted towards the listening position.

[0031] The addition of a single waveguide has the surprising effect of blocking the propagation of high-frequency side lobes, greatly limiting crosstalk for the beam of sound waves reflected further forward relative to the listening position.

[0032] Against all expectations, the invention then makes it possible to meet the criteria of Dolby Atmos® 5.1.4 or 7.1.4 formats by going beyond the angles usually observed, that is to say with an angle of inclination between the axis of revolution of the loudspeakers and the reflection surface, between 30° and 70°, while providing a compact speaker.

[0033] Accordingly to a first aspect, the invention relates to a reflective sound diffusion enclosure intended to diffuse sound waves in a space comprising at least one flat reflective surface, forming a plane of reflection for said sound waves; the reflective sound diffusion enclosure comprising a chassis which includes a mounting surface for reflective sound diffusion loudspeakers; an array of at least two loudspeakers juxtaposed and mounted on said mounting surface of the chassis, each loudspeaker comprising an axis of revolution and is arranged with said axis of revolution forming an angle of inclination with respect to the plane of reflection; and a control system configured to send a specific electrical signal to each loudspeaker and to control the electrical signals in amplitude and phase according to a spatial filtering method.

[0034] The reflective sound diffuser is remarkable in that it also includes a waveguide mounted on the chassis; the waveguide comprising at least one acoustic wall positioned in front of the loudspeakers; and in that each loudspeaker has an inclination angle between 30° and 70°. With the significant inclination angle of the loudspeakers, the invention offers a solution suitable for a wide range of rooms, particularly rooms where the listening position is located more than two meters from the diffuser.

[0035] Furthermore, the cooperation between the waveguide and the spatial filtering method makes it possible to limit crosstalk, even with a small footprint.

[0036] Indeed, the speaker array can be made with small drivers to fit into a soundbar-type enclosure. The limited directivity of the beam produced by these small drivers at low frequencies is compensated for by the waveguide. The combination of the waveguide and spatial filtering increases the speaker's directivity at low frequencies. At high frequencies, the waveguide acts as an acoustic barrier, limiting the direct propagation of secondary beams towards the listener. In one embodiment, the signal sent to the drivers comprises a first component generated using a spatial filtering method to obtain a first beam of sound waves, and a second component generated using a spatial filtering method to obtain a second beam of sound waves with a different directivity than the first beam of sound waves.

[0037] Thus, the reflective sound diffusion enclosure can emit two distinct beams from the speakers.

[0038] One beam can generate a phantom speaker above and in front of the listener, while another beam can generate a phantom speaker above and further behind the listener to reproduce the effects of a Dolby Atmos® 5.1.4 or 7.1.4 format.

[0039] Advantageously, the control system includes a filter to generate a different signal for at least one of the loudspeakers. The filter can be an analog electronic filter, a passive filter, or a digital filter. This makes it possible to limit the diffusion of high frequencies for the loudspeaker furthest from the acoustic wall of the waveguide.

[0040] Preferably, the control system applies to at least one loudspeaker, a low-pass type filter with a cutoff frequency between 5 kHz and 20 kHz, preferably 10 kHz.

[0041] In some embodiments, the control system applies a digital filter to at least one loudspeaker, calculated using a method known as "Pressure Matching." This method uses measurements or simulations of the acoustic pressure generated by each loudspeaker at predefined points in space to create filters that produce a beam with predefined properties, notably by canceling the acoustic pressure in a given direction. This helps to limit the level of crosstalk.

[0042] In some embodiments, the loudspeakers are arranged with a distance of less than 8 cm between the respective edges of two adjacent loudspeakers. This distance allows for a particularly effective beam emission angle through spatial filtering.

[0043] Preferably, the acoustic wall of the waveguide should be at least 4 cm high. The waveguide then very effectively limits the emission of sound waves, particularly at low frequencies.

[0044] In some embodiments, the loudspeakers are inclined along a plane of inclination, and the acoustic wall of the waveguide forms an angle between 50° and 100° with the plane of inclination of the loudspeakers.

[0045] Advantageously, the waveguide has an elliptical or rounded profile relative to the loudspeakers. An elliptical or rounded profile helps to limit wave diffraction and improve the waveguide's performance.

[0046] In some embodiments, the loudspeakers are arranged in different parallel planes. This loudspeaker configuration facilitates the formation of a sound wave beam and allows control over the orientation of the sound wave beam.

[0047] According to a second aspect, the invention relates to a method for generating spatialized sound comprising the steps of: providing a home cinema installation comprising at least one sound diffusion speaker by reflection according to the first aspect; sending to each speaker of the speaker a first signal component, according to a spatial filtering method to generate a first wide beam of sound waves; sending to each speaker of the speaker a second signal component, according to a spatial filtering method to generate a second directional beam of sound waves.

[0048] According to a third aspect, the invention relates to a home cinema installation comprising at least one sound diffusion speaker by reflection according to the first aspect.

[0049] BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The invention will be well understood and other aspects and advantages will become clear upon reading the following description, given by way of example with reference to the attached plates of figures on which:

[0051] [Fig. 1] is a schematic cross-sectional view of a state-of-the-art reflective sound diffusion enclosure implemented in a home cinema installation;

[0052] [fig. 2] is a schematic perspective view of a sound diffusion enclosure by reflection, according to an embodiment of the invention, implemented in a home cinema installation;

[0053] [fig. 3] is a schematic cross-sectional view of a loudspeaker and waveguide of the reflective sound diffusion enclosure along the cross-sectional plane Pc of figure 2;

[0054] [Fig. 4] is a schematic cross-sectional view, as in Figure 3, of a loudspeaker and waveguide of the reflective sound-diffusion enclosure according to another embodiment of the invention; [Fig. 5] is a schematic cross-sectional view of a reflective sound-diffusion enclosure, according to another embodiment of the invention, implemented in a home cinema installation; and

[0055] [fig. 6] is an explanatory diagram of the signal generated by the sound diffusion system by reflection in figure 5.

[0056] DETAILED DESCRIPTION OF THE INVENTION

[0057] An embodiment of a sound-diffusion enclosure 20 using reflection is illustrated in Figures 2 and 3. The enclosure 20 can be integrated into a listening room 22 or a home theater, for example, a room dedicated to home cinema activities, or even a living room. The enclosure 20 is designed to diffuse sound waves in a space comprising at least one substantially flat reflective surface, forming a reflection plane 24 for said sound waves. The reflective surface can be formed on a wall, a deflector, or any other object capable of reflecting a sound wave.

[0058] The listening room 22 comprises a ceiling, which here forms the reflection plane 24, a floor 28, and side walls. Typically, a listening room has four side walls, and the height from the ceiling to the floor is approximately 2.5 m. Of course, the invention can be adapted for a room with a different ceiling height and more or fewer side walls.

[0059] The sound-reflecting enclosure 20 is positioned on the floor 28 at the level of a front wall 30 and faces a listening position Pe at which a listener, not shown, can be installed.

[0060] It is possible to provide several listening positions within a listening area, and the listening area can be more or less large depending on the desired configuration. For example, the distance between the speaker 20 and the listening point Pe can be between 2 and 4 meters. The speaker 20 has a chassis 32 which includes a mounting surface 34 for reflective sound-diffusion loudspeakers 36. The mounting surface 34 is preferably flat but may include curves or angles to conform to the speaker's aesthetic appearance.

[0061] According to the invention, the enclosure 20 comprises an array of at least two loudspeakers 36 placed side by side and mounted on the mounting surface 34 of the chassis 32. As illustrated in Figure 3, each loudspeaker 36 has an axis of revolution Ar and is arranged with an axis of revolution Ar forming an angle of inclination P with respect to the reflection plane 24. For simplicity, the angle of inclination is shown with respect to a plane parallel Ppp to the reflection plane 24. Preferably, the angle of inclination P is fixed. Adding a system for varying the angle of inclination P manually or electronically increases the complexity of the enclosure and is not necessary for its operation. Preferably, the loudspeakers 36 are arranged on the same plane.The angle of inclination P can be measured with respect to the plane of the speaker in the case of a substantially flat speaker, or with respect to a plane perpendicular to a generatory line of the speaker cone in the case of a conical speaker.

[0062] According to the invention, the tilt angle P of the loudspeakers is between 30° and 70°; preferably, the tilt angle P is between 30° and 69°, the tilt angle P is between 45° and 69°. Preferably the tilt angle P is 55°.

[0063] In embodiments, as illustrated in Figure 4, the loudspeakers 36 can be arranged in different, parallel, planes Php, i.e., planes that do not coincide. The plane Php corresponds to the plane of the loudspeaker in the case of a substantially flat loudspeaker, or to a plane perpendicular to a line generating the loudspeaker cone in the case of a conical loudspeaker. This configuration allows for beam forming, oriented in a specific direction. Each loudspeaker emits an acoustic pressure directed along the axis Ar normal to the surface of the loudspeaker diaphragm, but the combination of the waves produces a main pressure lobe inclined at an angle G with respect to Ar. The direction of the main lobe is represented by a dashed arrow in Figure 4. Preferably, the angle G is between 0° and ±30°.

[0064] A person skilled in the art will understand how to position the Php planes to obtain a positive or negative angle G.

[0065] The distance between the speaker planes and the distance between the speaker axes Ar define the beam tilt angle G. A person skilled in the art will be able to calculate the distance between the planes Php to obtain the desired angle G. For example, the distance between the planes Php can be less than 10 cm; preferably less than 5 cm; preferably less than 1 cm.

[0066] According to variants not shown, the Php planes can form a non-zero angle with the mounting surface 34. Preferably, the Php planes can form an angle between 0° and ±30° with the mounting surface 34. Preferably, the Php planes can form an angle between 0° and ±20° with the mounting surface 34.

[0067] These variations make it easier to form a beam of sound waves.

[0068] According to unrepresented variants, the Php planes can be inclined relative to each other.

[0069] A control system 42 installed in the enclosure is configured to send a specific electrical signal to each loudspeaker 36 and to control the electrical signals in amplitude and phase according to a spatial filtering method.

[0070] Within the scope of the invention, a spatial filtering method is a signal processing method for forming a beam from an audio signal. It is implemented by combining an array of at least two loudspeakers individually controlled in phase and amplitude such that, in certain directions, the loudspeaker signals interfere constructively, while in other directions the signals interfere destructively. Several spatial filtering methods can be used within the scope of the invention, depending on the implementation requirements.

[0071] The Speaker 20 is ideally a compact speaker that can be placed on a piece of furniture such as a hi-fi cabinet or a living room TV stand. For example, the Speaker 20 is a soundbar that can be positioned under a screen. Therefore, the Speaker 20 has small dimensions, for example, a height of less than 25 cm and a depth of less than 40 cm.

[0072] The mounting surface 34 can be arranged on a top face of the enclosure, as illustrated in the embodiment of Figure 2.

[0073] In other embodiments, the mounting surface can be arranged on another face of the enclosure, for example to use a wall as a reflective surface.

[0074] The enclosure 20 comprises an array of at least two speakers 36 juxtaposed and mounted on the mounting surface 34.

[0075] Preferably, the loudspeakers 36 are structurally identical, but they may have different shapes or be made of different materials without altering the invention. In some embodiments, the loudspeaker diaphragms are advantageously substantially oval in shape, also known as "Race Track" diaphragms, in order to reduce the distance between the centers of each loudspeaker. For example, the loudspeakers may include at least one midrange or full-range driver and at least one tweeter.

[0076] The loudspeakers in the enclosure 20 are preferably aligned along the same axis, or along orthogonal axes, but other loudspeaker configurations enabling spatial filtering are conceivable. The enclosure 20 further includes a waveguide 38 mounted on the chassis 32; the waveguide 38 comprising at least one acoustic wall 40 positioned in front of the loudspeakers 36. The waveguide 38 may include several acoustic walls. For example, the waveguide may have a conical or pyramidal shape.

[0077] The acoustic wall 40 is positioned in front of the loudspeakers 36 in the direction of the listening position Pe. The acoustic wall 40 can then limit the transmission of sound waves along a direct line between the loudspeakers 36 and the listening position Pe.

[0078] The characteristics of the acoustic wall 40 are then appreciated in a cross-section plane Pc which includes a direct line, represented by Sd in Figure 2, between one of the loudspeakers and the listening position Pe.

[0079] In the Pc plane, the acoustic wall 40 may include a linear profile opposite the loudspeakers 36, as illustrated in Figure 3. Alternatively, the acoustic wall may have a profile including angles, or a convex profile, for example elliptical or rounded, opposite the loudspeakers.

[0080] As illustrated in Figure 3, the acoustic wall 40 of the waveguide has a height h of at least 2 cm, preferably at least 4 cm, and preferably between 4 cm and 20 cm. In the example in Figures 2 and 3, the height h coincides with the vertical, and the cross-sectional plane is that of Figures 2 and 3. The height h depends on the size, number, and spacing between the loudspeakers in the plane Pc.

[0081] As illustrated in Figure 3, the distance d between the edge of the two juxtaposed loudspeakers 36 is preferably less than 10 cm, preferably less than 8 cm, preferably the loudspeakers are contiguous.

[0082] In the Pc plane, the acoustic wall 40 of the waveguide 38 forms an angle 9 with the plane of inclination of the loudspeakers 36. The angle 9 depends on the height and profile of the acoustic wall. Preferably, the angle 9 is fixed. As with the inclination angle P, adding a system to vary the angle 9 manually or electronically would increase the complexity of the enclosure and is not necessary for its operation. The angle 9 is preferably between 45° and 110°, preferably between 70° and 100°.

[0083] In another embodiment not shown, the speaker is configured to diffuse sound by reflection off a vertical reflective surface, for example, a side wall of a room. In this embodiment, the tilt angle of the loudspeakers is also between 30° and 70°; preferably, the tilt angle is between 30° and 55°. Preferably, the tilt angle of the loudspeakers is 45°.

[0084] To simplify the explanation of the generation of sound wave beams by the control system 42, Figure 5 illustrates an array of two loudspeakers 36. Of course, it is possible to use a system with an array of more than two loudspeakers.

[0085] The control system 42 can be configured to apply a phase delay (Ati to AU) or an amplitude delay in a specific, i.e., independent manner, to the signals sent to the loudspeakers 36.

[0086] Within the framework of the invention, signal filtering can be applied with passive filters or with their equivalent in digital filter, for example with a digital signal processor, infinite impulse response filters, or finite impulse response filters.

[0087] As illustrated in figures 4 and 5; the control system 42 can also be configured so that the speaker 29 emits two different beams of sound waves towards the listening position Pe.

[0088] For this, the signal sent to the speakers includes a first component Ui(t) generated according to a spatial filtering method in order to obtain a first beam of sound waves Fl, and a second component U2(t) generated according to a spatial filtering method in order to obtain a second beam of sound waves F2, with a different directivity from the first beam of sound waves.

[0089] As illustrated in Figure 5, the first beam Fl can then reach the listening position Pe after a reflection off the ceiling 26, thus creating a phantom speaker further forward of the listening position Pe. The second beam F2 can reach the listening position Pe after one or two reflections, for example off the ceiling 26 and a rear wall 44, thus creating a phantom speaker further back from the listening position Pe.

[0090] The control system 42 may include a filter to generate a different signal for at least one of the speakers.

[0091] For example, as illustrated in Figure 6, a low-pass filter (LPF) can be applied to the signal sent to the loudspeaker furthest from the acoustic wall of the waveguide. The filter's cutoff frequency can be, for example, between 5 kHz and 20 kHz, preferably at 10 kHz.

[0092] In a preferred embodiment illustrated in Figure 6, it is possible to add a "High Shelf" (FHS) filter that amplifies or attenuates frequencies above the cutoff frequency by a specified amount to the loudspeaker closest to the acoustic wall of the waveguide, in order to increase its energy and maintain the total high-frequency energy.

[0093] The control system 42 may also include series filtering with, for example, a capacitor connected to the terminals of the loudspeaker 36. For example, the capacitance of the capacitor has a value between 1pF and 20pF, preferably between 2pF and 12pF, preferably with a value of 6.8pF.

[0094] In this example, lowering the level of loudspeaker 36 reduces the effect of spatial filtering, thus restoring a directivity close to that of a single directional loudspeaker at very high frequencies. This results in a wider directivity, providing a larger and more homogeneous listening area, i.e., with fewer side lobes. In another, unshown, embodiment, the beam direction is achieved by modifying the relative phase of the loudspeakers, for example, by using one or more all-pass filters.

[0095] In another, unshown embodiment, a digital filter calculated using a method known as "Pressure Maching" can be employed. This method uses measurements or simulations of the acoustic pressure generated by each loudspeaker at predefined points in space to create filters that produce a beam with predefined properties, notably by canceling the acoustic pressure in a given direction. This helps to limit the level of crosstalk.

[0096] A process for generating spatialized sound may include the following steps:

[0097] - to provide a home cinema system comprising at least one 20-inch reflecting sound speaker according to the invention;

[0098] - send to each speaker 36 of the enclosure a first signal component Ui(t), according to a spatial filtering method to generate a first wide beam of sound waves Fl;

[0099] - send to each speaker 36 of the enclosure a second signal component U2(t), according to a spatial filtering method to generate a second directional sound wave beam F2.

[0100] In some embodiments, the method may further include a step of applying a filter to the first signal component Ui(t) or to the second signal component U2(t), or to the sum of the components Ui(t) + U2(t) of a loudspeaker. For example, the filter may be a low-pass filter or an all-pass filter.

[0101] With this method, it is then possible to generate phantom speakers allowing to reproduce sound effects of a Dolby Atmos® 5.1.4 or 7.1.4 format. It should be noted that, although the process of generating spatialized sound has been described in a certain order, it is possible to use a different ordering of the steps.

[0102] In conclusion, the invention makes it possible to obtain a speaker 20 usable for compact systems, with limited crosstalk and offering better flexibility in listener positioning.

Claims

DEMANDS 1. A reflective sound-diffusion enclosure (20) intended to diffuse sound waves in a space comprising at least one flat reflective surface, forming a plane of reflection for said sound waves; the reflective sound-diffusion enclosure comprising: - a chassis (32) which includes a mounting surface (34) for sound-reflecting loudspeakers (36); - an array of at least two loudspeakers (36) placed side by side and mounted on said mounting surface (34) of the chassis, each loudspeaker (36) comprising an axis of revolution (Ar) and arranged with said axis of revolution (Ar) forming an angle of inclination (P) with respect to the reflection plane (24); and - a control system (42) configured to send a specific electrical signal to each loudspeaker (36) and to control the electrical signals in amplitude and phase according to a spatial filtering method; characterized in that the enclosure (20) further comprises a waveguide (38) mounted on the chassis (32); the waveguide (38) comprising at least one acoustic wall (40) disposed in front of the loudspeakers (36); and in that each loudspeaker (36) has an angle of inclination between 30° and 70°.

2. Sound diffusion enclosure by reflection according to claim 1, in which the signal sent to the loudspeakers (36) comprises a first component Ui(t) generated according to a spatial filtering method so as to obtain a first beam of sound waves (Fl), and a second component U2(t) generated according to a spatial filtering method so as to obtain a second beam of sound waves (F2), with a different directivity from the first beam of sound waves (Fl).

3. Sound diffusion enclosure by reflection according to claim 2, wherein the control system includes at least one filter for generating a different signal for at least one of the loudspeakers 36.

4. Sound diffusion enclosure by reflection according to claim 3, in which the control system applies to at least one loudspeaker (36), a low-pass type filter with a cutoff frequency between 5 kHz and 20 kHz, preferably 10 kHz.

5. Sound diffusion enclosure by reflection according to any one of the preceding claims, wherein the control system applies to at least one loudspeaker (36), a digital filter calculated from the method, known in English as “Pressure Matching”.

6. Sound diffusion enclosure by reflection according to any one of the preceding claims, in which the loudspeakers (36) are arranged with a distance d of less than 8 cm between a respective edge of two juxtaposed loudspeakers (36).

7. Sound diffusion enclosure by reflection according to any one of the preceding claims, in which the acoustic wall (40) of the waveguide (38) has a height h of at least 4 cm.

8. Sound diffusion enclosure by reflection according to any one of the preceding claims, the loudspeakers (36) being inclined along a plane of inclination, in which the acoustic wall (40) of the waveguide (38) forms an angle 9 between 50° and 100° with the plane of inclination of the loudspeakers (36).

9. Sound diffusion enclosure by reflection according to any one of the preceding claims, in which the waveguide (38) has an elliptical or rounded profile with respect to the loudspeakers (36).

10. Sound diffusion enclosure by reflection according to any one of the preceding claims, in which the loudspeakers (36) are arranged in different parallel planes (Php).

11. Method for generating spatialized sound comprising the steps of: - providing a home cinema installation comprising at least one speaker (20) for sound diffusion by reflection according to one of the preceding claims; - send to each speaker (36) of the enclosure a first signal component Ui(t), according to a spatial filtering method to generate a first wide beam of sound waves (Fl); - send to each speaker (36) of the enclosure a second signal component U2(t), according to a spatial filtering method to generate a second directional sound wave beam (F2).

12. Home cinema installation comprising at least one reflecting sound speaker (20) according to any one of claims 1 to 10.

Images