Device for sound detection, sound localization and beamforming and method for manufacturing the same

Abstract

A device (34) for sound detection, sound localization and beamforming, comprising a display (1) and a plurality of microphone stacks (3), wherein the display (1) surrounds each microphone stack (3) in a transverse direction (x, y). The device (34) further comprises a plurality of elastic connectors (7), wherein each elastic connector (7) surrounds a corresponding microphone stack (3) in a transverse direction (x, y) and mechanically connects the corresponding microphone stack (3) to the display (1). Each microphone stack (3) further comprises a microelectromechanical transducer array (5), wherein the transducer array (5) comprises a plurality of membranes (14), in particular nano-membranes, and a corresponding integrated back volume (17), which is arranged below the membrane (14). An optical reader (6) is configured to detect the displacement of each membrane (14) individually.

Description

[0001] The present disclosure relates to an apparatus for sound detection, sound localization, and beamforming, an electronic device, and a method for manufacturing the apparatus for sound detection, sound localization, and beamforming.

[0002] Microphone arrays with discrete microphones have been used in various applications, such as to achieve directionality, sound localization, and beamforming. A variety of approaches are available for sound localization and beamforming, ranging from evaluating cross-correlation and autocorrelation functions to delay-and-sum or track-while-scanning algorithms, taking into account both far-field and near-field acoustic signals.

[0003] To achieve good spatial resolution, an ideal microphone array should have a large aperture (the size of the entire array), since the Rayleigh criterion for resolution states that θ = sin -1 (1.22·λ / D), where θ is the minimum angle between two sources to be detected, D is the size of the linear aperture, and λ is the wavelength of the sound pressure wave. This means that small apertures do not have good directivity at lower frequencies.

[0004] On the other hand, an array with a large aperture and a small number of transducers suffers from a loss of directivity at higher frequencies, which is known as sidelobes. For a one-dimensional array with N equally spaced transducers, this can be derived from the spatial Fourier transform of the array's aperture function (also called the mode function). In this case, the mode function can be written as:

[0005]

[0006] where k = 2π / λ = ω / C0 is the acoustic wave number associated with a plane wave with phase velocity c0, d is the distance between adjacent transducers, λ is the wavelength, and is the angle at which the wave strikes the one-dimensional transducer array. The mode function w(k) is a periodic function of k. In addition to the main lobe at k = 0, the mode function also exhibits repetition of the main lobe at positions k = p·(2π / d) (where p = ±1, ±2…). Assuming d is less than half the wavelength of the acoustic wave, the directional response of the array is dominated by the main lobe. However, once d exceeds λ / 2, "spatial aliasing" occurs and significant side lobes are generated in the response of the array. This means that in order to achieve the maximum frequency requirements, the distance between the transducers in the array must be less than half the wavelength of the acoustic wave at that frequency. However, these requirements require a large aperture and a large number of transducers.

[0007] For beamforming algorithms, it should be noted that if appropriate weighting functions are applied to the transducer arrays to ensure that their net output is proportional to the angle of arrival, coherently add, the transducer array can be steered to The sound of the impact is selective. In this case, the mode function can be written as:

[0008]

[0009] This function corresponds to a shift of the spatial Fourier transform, resulting in a main lobe with the desired angle As the center.

[0010] Capacitive microelectromechanical (MEMS) microphones have been used as discrete components to form sensor arrays that can have linear, rectangular, or circular shapes. However, such arrays require the use of individual microphones, which can be costly. Furthermore, capacitive MEMS microphones used as discrete components in an array require that each transducer be exposed to the sound field, which requires considerable space and volume. Due to the method of reading the membrane displacement, capacitive MEMS microphone arrays with integrated transducers require very complex powering and sensing schemes. Since each microphone must operate as an individual transducer in the array, the acoustic requirements of the back volume in each transducer can also occupy a considerable volume.

[0011] An object is to provide a device for sound detection, sound localization and beamforming that can be implemented in a reduced volume. Another object is to provide a method for manufacturing a device for sound detection, sound localization and beamforming that can be implemented in a reduced volume.

[0012] This object is achieved by the independent claims. Further embodiments and variants follow from the dependent claims. Unless otherwise stated, the above definitions also apply to the following description.

[0013] In one embodiment, an apparatus for sound detection, sound localization, and beamforming includes a display having a main plane of extension. The display can be the display of any electronic device, such as a mobile device (e.g., a cell phone). The display can comprise any known material for constructing such a display. The display can be transparent to electromagnetic radiation, at least in wavelengths of the visible spectrum. The display can be planar, i.e., not curved, relative to the main plane of extension of the display.

[0014] The device further comprises a plurality of microphone stacks, wherein the display surrounds each microphone stack in a lateral direction. The lateral direction extends parallel to the main extension plane of the display. This means that the respective microphone stack is inserted into a gap within the display. However, the gap separates the display and the inserted microphone stack. Therefore, the display and the microphone stack have no direct mechanical contact. The number of microphone stacks is arbitrary and depends on the application. For example, the device comprises at least two microphone stacks.

[0015] The device also includes a plurality of elastic connectors, each of which surrounds a corresponding microphone stack in the lateral direction and mechanically connects the corresponding microphone stack to the display. This means that each elastic connector is in mechanical contact with both the display and the corresponding microphone stack. Each elastic connector can completely surround a corresponding microphone stack in the lateral direction. Each elastic connector can comprise any material having a lower Young's modulus value than the Young's modulus of the material used for the display. This means that the hardness of the elastic material is lower than the hardness of the material of the display. For example, the elastic layer can comprise silicone. The elastic layer can form a ring that spans the gap between the display and the microphone stack.

[0016] The microphone stack comprises a microelectromechanical transducer array comprising a plurality of membranes and a corresponding integrated rear volume. The transducer array extends in a direction parallel to a main extension plane of the display. The membranes can be integrated on a substrate, in particular a semiconductor substrate, for example a semiconductor substrate comprising silicon (Si). In addition, the membranes can comprise materials compatible with semiconductor processes, such as silicon nitride (SiN) or polycrystalline silicon. However, the membranes can also comprise two-dimensional materials (2D materials), such as graphene or molybdenum disulfide (MoS2) or any other material suitable for forming nanofilms. The two-dimensional material should have a thickness of a single atomic layer. However, other thicknesses are also conceivable. In one embodiment, the thickness of the 2D material is two atomic layers. However, a larger number of atomic layers is also feasible, for example 10 layers.

[0017] The rear volume is disposed transversely below the membrane. This transverse direction extends perpendicular to the main extension plane of the display. This means that the rear volume is formed by the volume below the membrane. The rear volume can also be formed by a cavity in the substrate supporting the membrane. The rear volume can be filled with air or gas.

[0018] The MEMS transducer array comprises at least two MEMS transducers arranged in a regular or irregular pattern relative to one another. For example, four transducers can be arranged in a rectangular grid. Each transducer is configured to convert an acoustic pressure wave into a mechanical deflection of the membrane, which can in turn be converted into an electrical signal by a readout method. The rear volume is configured to provide acoustic compliance. The response of the membrane to the acoustic pressure waves, particularly at audio frequencies, depends on the size of the rear volume. This means that the rear volume should be as large as possible to improve the sensitivity and signal-to-noise ratio (SNR) of the transducer.

[0019] The microphone stack also includes an optical pickup device that is configured to individually detect the displacement of each membrane within the transducer array. This means that each transducer within the transducer array forms an optical microphone. Optical microphones evaluate the response of the membrane to sound by using optical methods. For example, the optical pickup device can include at least one light source and a photodetector. The light source can be a laser, an LED, or other light source with a controlled coherence length. When the reflector is integrated on the membrane, vibrations can be detected by using a laser beam and an interferometer, i.e., the movement of the membrane can be detected by changes in the optical path length of the reflected laser beam. The interferometer can be used to convert the associated phase changes into intensity changes of the light signal. The optical pickup device detects the displacement of each transducer within the transducer array separately.

[0020] Due to the small size of MEMS transducer arrays, they can be installed in devices with limited size and volume, such as mobile devices and smartphone displays. Optical readout allows the transducers to be small in the lateral direction and thin in height. The integrated rear volume provides sufficient acoustic compliance even for small-diameter membranes.

[0021] Even with a single transducer array, good directivity and spatial resolution of incoming sound waves can be achieved at the lowest possible audio frequencies (preferably down to 10 kHz). Furthermore, using multiple membranes in a transducer array improves the SNR by a factor of 10·log(N), where N is the number of membranes in the transducer array.

[0022] By combining more than one transducer array, sound localization and beamforming are enabled even at low frequencies (preferably down to 1 kHz).

[0023] Each membrane is directly exposed to the acoustic pressure by using elastic connectors to mount the transducer array. The elastic connectors reduce the structure-borne sound transmission from the display to the transducer array.

[0024] In one embodiment, each microphone stack further comprises an elastic layer. The elastic layer can comprise any material having a lower Young's modulus value than the Young's modulus of the material used for the display. For example, the elastic layer can comprise a nanomaterial. For example, the elastic layer can be carbon and form a carbon nanofilm. In another example, the nanofilm can be made of molybdenum disulfide. Because the elastic layer is the topmost layer of the microphone stack, it is connected to the environment above the display. The elastic layer can be strong so as to withstand environmental influences and external forces, such as touch. However, the elastic layer also acts as a membrane and can deflect in the transverse direction. In this way, the elastic layer transmits sound pressure waves toward the transducer array.

[0025] In this embodiment, each microphone stack further comprises a gasket, wherein the gasket is arranged transversely between the elastic layer and the transducer array. The gasket has a first surface and a second surface opposite the first surface. The gasket is in mechanical contact with the transducer array via the first surface and in mechanical contact with the elastic layer via the second surface. In addition, the gasket comprises a plurality of through-holes penetrating the gasket. This means that the through-holes connect the first surface and the second surface of the gasket. The membrane of the transducer array is aligned with the through-holes of the gasket. The alignment is achieved such that the through-holes are located above the membrane in the transverse direction. Thus, the gasket separates the sound pressure of each membrane. The volume enclosed by the membrane, the through-holes, and the elastic layer is called the front volume. The front volume can be filled with air or gas and sealed by the gasket. The gasket can comprise a material that is as hard as possible in terms of Young's modulus. For example, the gasket comprises silicon or glass.

[0026] By providing an elastic layer and spacers above the transducer array, the sensitive membranes of the transducer array are protected from environmental influences and external forces, such as touch. This is because the membranes are located within the device and separated from the display by the spacers. Due to the flexibility of the elastic layer and its membrane function above each through-hole, sound pressure waves striking the elastic layer are able to propagate forward toward the membranes in the transducer array. This protective arrangement allows the transducer array to be manufactured using conventional MEMS technology and standard MEMS materials. For example, the membrane comprises SiN.

[0027] In another embodiment, each microphone stack further comprises a 2D material layer as the topmost layer, wherein the membrane of the transducer array is formed by a corresponding portion of the 2D material layer. This means that each membrane comprises a 2D material. In this embodiment, the membrane is directly exposed to the environment above the display. Therefore, no elastic layer or gasket is required. For example, the 2D material layer can comprise carbon (resulting in graphene when the thickness is one atomic layer) or molybdenum disulfide (MoS2). However, any other material suitable for forming a nanomembrane can be used. The 2D material layer can comprise several atomic layers of the corresponding material. For example, the 2D material layer can be thin and comprise only one or two atomic layers. However, thick 2D material layers corresponding to tens of atomic layers are also possible. The total thickness, i.e. the number of atomic layers, is arbitrary and depends on the application. The corresponding material, thickness and geometric arrangement of the 2D material layer influence the dependence of the membrane displacement on pressure.

[0028] By using 2D materials (such as graphene or MoS2) for the membrane and due to the properties of 2D materials (such as Young's modulus, density, etc.), a smaller membrane diameter can be achieved. Therefore, even if the aperture of the transducer array is small, the membranes can have a larger spacing. In this way, even a single transducer array can have good directivity and spatial resolution at low frequencies (for example, down to 10kHz). Membranes containing 2D materials are more resistant to environmental influences, so they can be directly exposed to the environment.

[0029] In one embodiment, each transducer array further includes a substrate disposed transversely below the plurality of membranes. The substrate may comprise a semiconductor material, such as Si. The membranes may be disposed on top of the substrate, and the membranes may be formed from portions of the membranes. A cavity is formed in the substrate beneath corresponding portions of the membranes, exposing the membranes on a side facing the substrate. The exposed portions of the membranes form the membranes. The membranes may be planar or corrugated.

[0030] The rear volumes are formed by cavities in the substrate beneath the respective membranes. Thus, the rear volumes are integrated into the substrate. Each rear volume can be filled with a gas or air. For example, the rear volumes can be filled with nitrogen.

[0031] By using a substrate, particularly a semiconductor substrate, the transducer device can be manufactured using conventional processes such as deposition and etching techniques. Furthermore, additional functional components, such as circuitry, sensors, and electrical connectors, can be integrated into the substrate. Because the rear volume is integrated into the substrate, the transducer array can be very compact, allowing for further miniaturization of the device.

[0032] In one embodiment, the transducer array further includes vents connecting each rear volume to the environment. These vents are formed by openings in the substrate. These openings can be arranged adjacent to each other in a lateral direction extending parallel to the main extension plane of the display. In this way, each opening opens onto a side surface of the substrate. The openings can be narrow to form a channel. For each rear volume, there can be a corresponding vent connecting the rear volume to the environment.

[0033] The advantage of this structure is that the acoustic resistance of the vent can be increased arbitrarily. For narrow and long openings, the acoustic resistance is very high. By increasing the acoustic resistance, pressure drops are avoided and the noise level of the device can be reduced.

[0034] Alternatively, a vent can be integrated into each membrane. For example, the vent can be located in the center of the membrane relative to its lateral extent. This means that the vent penetrates the membrane. If a gasket and elastic layer are present, another vent can be located in the elastic layer above each membrane, where this additional vent penetrates the elastic layer. In this way, each rear volume is connected to the environment via a vent in the membrane, the front volume, and another vent in the elastic layer. The vent can have a small diameter compared to the membrane diameter so as not to adversely affect the function and sensitivity of the membrane.

[0035] The vent provides pressure equalization between the rear volume and the environment. This may be necessary because changes in gas density within the rear volume due to membrane deflection can affect system compliance and, therefore, the membrane's dynamic properties. On a timescale comparable to the signal period of the acoustic pressure wave, pressure equalization between the rear volume and the environment is slow.

[0036] In one embodiment, the optical pickup device comprises at least one light source and at least one photodetector, wherein the light source and the photodetector are arranged on a further substrate, wherein the further substrate is attached to the transducer array below the transducer array in a transverse direction.

[0037] The other substrate may also include a semiconductor material, such as Si. A light source and a photodetector may be integrated on the other substrate. The light source may be, for example, a vertical cavity surface emitting laser (VCSEL). However, a different light source capable of providing coherent light of any wavelength may also be used. In addition, more than one light source may be integrated on the other substrate. The photodetector may be, for example, a photodiode or a photodiode array. However, different photodetectors may also be used. In addition, more than one photodetector may be integrated on the other substrate.

[0038] Additional components can also be integrated onto this other substrate. For example, circuits for triggering and evaluating optical signals, as well as interferometric devices such as mirrors, can also be arranged on the other substrate. As mentioned above, the vibration of the membrane can be detected using a laser beam from a light source. This is because the deflection of the membrane causes the optical path length of the reflected laser beam to change. This can detect changes in the phase of the light, or, when using an interferometer, changes in the intensity of the light.

[0039] The other substrate can be attached to the transducer array using any technique known in the art. For example, bonding techniques can be employed. Alternatively, the other substrate can be attached to the substrate comprising the transducer array using an adhesive. The other substrate is attached to the transducer array so that the displacement of each membrane within the transducer array can be individually detected. Light routing from the light source toward the membranes is provided by the respective rear volumes.

[0040] With an optical readout, the device can be smaller and thinner. This contrasts with capacitive MEMS microphones, which require complex power supply and sensing schemes due to their membrane displacement reading method. Consequently, capacitive MEMS microphones require more space.

[0041] In one embodiment, each transducer array has a rectangular shape in a top view. A top view refers to a view of the transducer array viewed from a side facing away from the optical pickup device along a transverse direction. In the transverse direction, the length of the transducer array is at least 1 mm and at most 10 mm.

[0042] In a preferred embodiment, the transducer arrays have a square shape in top view. In this preferred embodiment, the length of each transducer array is at least 1.5 mm and at most 5 mm. In a more preferred embodiment, each transducer array has a square shape in top view with a length in each lateral direction of at least 1.7 mm and at most 2.3 mm.

[0043] Because each transducer array has a small lateral extent, they can be installed into systems with limited space, such as smartphone displays. However, each transducer array must provide enough space for multiple membranes to provide good directivity and good SNR even at higher frequencies.

[0044] In one embodiment, each transducer array is spaced apart from an adjacent transducer array in the lateral direction by at least 1 cm and at most 10 cm.

[0045] In a preferred embodiment, each transducer array is spaced apart from an adjacent transducer array in the lateral direction by at least 3 cm and at most 7 cm.

[0046] By defining appropriate spacing between the transducer arrays, multiple transducer arrays can be arranged within the display. This allows for good directivity even at low frequencies. Such large spacing between transducer arrays is beneficial in terms of providing good spatial resolution during sound localization.

[0047] In one embodiment, each membrane within the transducer array has a circular shape in top view. The circular shape has a diameter of at least 0.1 mm and at most 1.5 mm.

[0048] Conventional materials like SiN can be used to construct membranes with larger diameters (i.e., about 1.5 mm in diameter). 2D materials can be used to achieve smaller membrane diameters, i.e., about 0.1 mm in diameter. Despite the small membrane diameter, such membranes still achieve acceptable acoustic performance due to the properties of 2D materials (such as Young's modulus and density): membranes containing 2D materials can have large deflections even for small diameters. Furthermore, such membranes can be very thin because they contain only a few atomic layers. Ideally, nanomembranes are less than 10 nm thick.

[0049] Defining the right diameter for each membrane influences the acoustic performance of each transducer. Furthermore, a smaller diameter allows for more membranes in a transducer array and a greater distance between them. Having more membranes results in better SNR and directivity at higher frequencies. A greater distance between membranes within a transducer array provides better directivity at lower frequencies.

[0050] In one embodiment, each membrane within the transducer array is spaced apart from adjacent membranes by a predetermined spacing. Spacing is defined as the distance between the centers of the respective membranes, where the center of a membrane refers to its lateral extent. The spacing is at least 0.4 mm and at most 3 mm.

[0051] In a preferred embodiment, the spacing is at least 0.4 mm and at most 1.7 mm. As described above, smaller spacing allows more membranes to be placed in a single transducer array, which in turn increases the SNR and enhances directivity at higher frequencies. Larger spacing between membranes within the transducer array enhances the directivity and spatial resolution of a single transducer array at lower frequencies.

[0052] In one embodiment, a device including a transducer array is incorporated into an electronic device. The electronic device can be, in particular, a smartphone, a smart speaker, a smart TV, a smart TV remote control, or any other household appliance or any other remote control. Since the transducer array can be manufactured using technology for microelectromechanical systems (MEMS) and can have a reduced size, the electronic device can have a small size.

[0053] Furthermore, a method of manufacturing an apparatus for sound detection, sound localization and beamforming is provided. All features disclosed for the apparatus are also disclosed for the method for manufacturing the apparatus, and vice versa.

[0054] A method for manufacturing an apparatus for sound detection, sound localization, and beamforming includes providing a display having a main extension plane. It also includes providing an optical readout device. The transducer array is manufactured by providing a substrate and forming a thin film on the top surface of the substrate. The thin film can comprise a single material. However, the thin film can also comprise a stack of different materials, including a sacrificial layer or an additional reflective layer that serves as a reflector for the optical readout method. The thin film can be formed by any technique known in the art. For example, the thin film can be formed by deposition, such as by chemical vapor deposition (CVD) or by plasma enhanced CVD (PECVD). In addition, other processing steps can be included to form the thin film. For example, the thin film can also be formed by a heat treatment step that causes solid-state dissolution. By applying a rapid thermal annealing (RTA) step, particles can be dissolved from one material layer into another material layer, thereby forming a new functional layer. In addition, the thin film can be patterned by an etching process.

[0055] After forming the thin film, grooves are formed in the substrate from the rear surface of the substrate towards the thin film. Grooves can be formed by deep reactive ion etching (DRIE) or by wet etching or by their combination. The etching process can be controlled by time or by using an etch stop layer. In the latter case, the etch stop layer has been deposited on the main surface of the substrate before forming the thin film, and the grooves should be formed. The etch stop layer can be removed by wet etching after forming the opening.

[0056] After forming the trench in the substrate, an optional sacrificial layer may need to be removed by an etching process to release the actual membrane. Multiple membranes are formed from the corresponding portions of the thin film exposed by etching the trench and the optional sacrificial layer. The corresponding integrated back volume is formed by the trench itself.

[0057] After the transducer array is manufactured, an optical pickup is mounted below the transducer array in a transverse direction, thereby forming a microphone stack. This transverse direction extends perpendicular to the main extension plane of the display. The mounting can be performed, for example, by bonding techniques or with the aid of adhesives. In particular, wafer-level bonding, such as anodic bonding, oxide-oxide bonding, or hybrid bonding, can be performed. The optical pickup is mounted so that the rear volume provides a light path towards the membrane for light from a light source on the optical pickup. Therefore, the optical pickup must be aligned with the transducer array.

[0058] In a next step, an elastic connector is formed around the microphone stack, the elastic connector surrounding the microphone stack in a lateral direction extending parallel to the main extension plane of the display. The elastic connector can be formed, for example, by molding.

[0059] The microphone stack is mounted in the display together with an elastic connector. This means that the corresponding microphone stack is inserted into a gap in the display. In landscape orientation, the display surrounds the microphone stack at a distance, with the elastic connector mechanically connecting the display to the microphone stack.

[0060] Well-known MEMS processes can be used to manufacture the transducer arrays. This manufacturing is very cost-effective. Furthermore, the optical pickup can also be manufactured using semiconductor processes. Due to the small size of each transducer array with its integrated back volume and the stacked arrangement with the optical pickup, such a microphone stack can be placed in the display of an electronic device where limited space is available.

[0061] In one embodiment of the method for manufacturing a microphone stack, the method further comprises providing an elastic layer. In addition, a gasket is provided having a plurality of through holes penetrating the gasket. If silicon is used as the material for the gasket, conventional processes such as DRIE can be applied to form the through holes. The gasket is mounted under the elastic layer in a transverse direction, for example with the aid of an adhesive. In addition, a transducer array with an optical reading device attached is mounted under the gasket and the elastic layer in a transverse direction. This means that the gasket is arranged between the elastic layer and the transducer array. The membrane of the transducer array is aligned with the through holes of the gasket. The mounting can be performed with the aid of an adhesive. In the case where the gasket comprises a Si or glass bonding process, in particular wafer-level bonding, such as anodic bonding or oxide-oxide bonding or hybrid bonding, can also be performed.

[0062] By providing an elastic layer and a spacer above the transducer array, the sensitive membrane of the transducer array is protected from environmental influences and external forces (such as touch). This is because the membrane is located within the device and is separated from the display by the spacer. However, due to the flexibility of the elastic layer, sound pressure waves can be transmitted to the membrane of the transducer array.

[0063] In another embodiment of the method, the membrane comprises a 2D material layer. The membrane of the transducer array is formed from corresponding portions of the 2D material layer. In this embodiment, the 2D material layer is the topmost layer of the microphone stack. Therefore, no elastic layer or spacer is required.

[0064] By using 2D materials for the membranes, and due to the properties of 2D materials such as Young's modulus, density, etc., smaller membrane diameters can be achieved. Membranes comprising 2D materials are more resistant to environmental influences, so they can be directly exposed to the environment.

[0065] The following description of the accompanying drawings further illustrates and explains the exemplary embodiments. Components having the same function or effect are denoted by the same reference numerals. Identical or substantially identical components may be described only for the figure in which they first appear, and their description need not be repeated in subsequent figures.

[0066] 1 a and 1 b show schematic diagrams of embodiments of a device comprising a display with two integrated microphone stacks.

[0067] Figure 2 An exploded assembly view of an embodiment of a microphone stack is shown.

[0068] 3a and 3b show details of an embodiment of a microphone stack including a vent.

[0069] 4a to 4c show schematic diagrams of two embodiments of a device comprising a display with four integrated microphone stacks.

[0070] 5a-5b show schematic diagrams of embodiments of a device comprising a display with six integrated microphone stacks.

[0071] Figure 6 A schematic diagram illustrating an exemplary embodiment of an electronic device including apparatus for sound detection, sound localization, and beamforming.

[0072] 7a to 7f illustrate an embodiment of a method of manufacturing an apparatus for sound detection, sound localization and beamforming.

[0073] 8a to 8f illustrate an embodiment of a method of fabricating a transducer array comprising two-dimensional (2D) material layers.

[0074] FIG1 a shows a perspective view of an apparatus 34 including a display 1. The display 1 can be included in any electronic device 20, such as a smartphone. The display 1 has a main extension plane extending parallel to the transverse directions x and y. Furthermore, the display 1 also extends along a transverse direction z perpendicular to the main extension plane of the display. The display has a top surface 2.

[0075] FIG1a also shows two microphone stacks 3 installed in the display 1. The display 1 surrounds each microphone stack 3 in the transverse directions x, y. There is a gap between the display 1 and each microphone stack 3. The microphone stack 3 has a further top surface 4. In the transverse direction z, the further top surface 4 of each microphone stack 3 and the top surface 2 of the display 1 are at the same level, so that they form a common surface. However, it is also possible that there is an offset between the top surface 2 of the display 1 and the further top surface 4 of the microphone stack 2 in the transverse direction z. Furthermore, in the transverse direction z, the microphone stack 3 comprises at least one transducer array 5 having a plurality of membranes 14 and corresponding rear volumes 17 (not shown) and an optical pickup device 6 (not shown). The transducer array 5 and the optical pickup device 6 are not shown in FIG1a.

[0076] The gap between the display 1 and the microphone stack 3 is filled by a corresponding elastic connector 7. Each elastic connector 7 surrounds a corresponding microphone stack 3 in the lateral directions x, y. In addition, each elastic connector 7 mechanically connects the display 1 to the microphone stack 3. For example, the elastic connector 7 can include PDMS.

[0077] FIG1 b shows a detail of a microphone stack 3 within a portion of a display 1. The microphone stack 3 has a rectangular shape when viewed from above, which refers to a view of the top surface 2 of the display 1 along the vertical direction z. However, in other embodiments, different shapes are possible. The extent of the microphone stack 3 in the transverse direction x is represented by the length l. The extent of the microphone stack 3 in the other transverse direction y is represented by the width w.

[0078] exist Figure 2 In FIG, an exploded view of an embodiment of a microphone stack 3 is shown. The different components of the microphone stack 3 are shown arranged relative to each other in the transverse direction z. Figure 2 In the illustrated embodiment, the microphone stack 3 comprises, from top to bottom, an elastic layer 8, a spacer 9, a transducer array 5, and an optical pickup 6. Furthermore, the microphone stack 3 can be disposed on a printed circuit board (PCB) 10 or a flexible PCB 10. All components of the microphone stack 3 can have the same measurements in a top view. This means that they all extend in the lateral directions x and y, with a width w and a length l.

[0079] Elastic layer 8 is the topmost layer of microphone stack 3. Elastic layer 8 can contain nanomaterials, such as carbon. Thus, it can form a carbon nanofilm. On the one hand, elastic layer 8 is flexible in terms of its Young's modulus. On the other hand, elastic layer 8 is resistant to environmental influences, thus protecting the underlying sensitive membrane 14 of transducer array 5.

[0080] A gasket 9 is disposed beneath the elastic layer 8 in the transverse direction z. The gasket has a first surface 11 and a second surface 12 opposite the first surface 11. The gasket 9 is in mechanical contact with the transducer array 5 via the first surface 11 and in mechanical contact with the elastic layer 8 via the second surface 12. In the transverse direction z, the gasket 9 extends according to a thickness t. A plurality of through-holes 13 penetrate the gasket 9. In this example, the gasket 9 includes nine through-holes 13 arranged in a regular 3×3 grid. These through-holes 13 penetrate the gasket 9 from its first surface 11 toward its second surface 12. Each through-hole has a diameter dg. The pitch of the through-holes 13, i.e., the distance between the centers of adjacent through-holes 13, is denoted by pg.

[0081] exist Figure 2 In the embodiment of FIG. 5 , the transducer array 5 includes nine membranes 14. Typically, the number of membranes 14 matches the number of through holes 13 in the gasket 9. Each membrane 14 has a circular shape in a top view. The diameter of each membrane 14 is denoted by dm. Figure 2 An embodiment is shown in which the transducer array 5 can be manufactured from conventional MEMS materials using conventional MEMS manufacturing techniques. This means that the transducer array 5 can comprise a semiconductor substrate 15 comprising, for example, Si. For example, the membranes can comprise SiN. Under each membrane 14 there is a corresponding back volume 17 formed by a cavity in the substrate 15. However, this cavity is not Figure 2 The plurality of membranes 14 is aligned with the plurality of through holes 13 of the gasket 9. This means that in the transverse direction z, above each membrane 14, a front volume 28 is formed by the corresponding through hole 13. The diameter dm of each membrane 14 can be equal to or smaller than the diameter dg of the through hole 13.

[0082] The optical pickup device 6 is arranged below the transducer array 5 along the transverse direction z. The optical pickup device 6 is attached to the transducer array 5 so that the displacement of each membrane 14 can be detected individually. The optical pickup device 6 can include another substrate 16, one or more light sources 25 (not shown), and one or more photodetectors 26 (not shown). In addition, the optical pickup device 6 can also be manufactured using semiconductor technology.

[0083] FIG3a shows a diagram according to Figure 2 FIG1 shows an embodiment of a microphone stack 3 without spacers 9 and elastic layer 8. In this figure, the rear volume 17 formed by the cavity in the substrate 15 is indicated by a dashed line for two of the nine membranes 14. The rear volume 17 is arranged below each membrane 14 in the transverse direction z. The height h of the rear volume 17, which corresponds to the extent of the rear volume 17 in the transverse direction z, depends on the height of the substrate 15.

[0084] FIG3 a further shows an exemplary vent hole 18 for one of the rear volumes 17. This vent hole 18 connects the corresponding rear volume 17 to the environment. In this example, the vent hole 18 is formed by an opening in the substrate 15. The vent hole 18 forms a channel extending in a transverse direction x from the rear volume 17 toward a side surface 19 of the substrate 15. Additional vent holes 18 for other rear volumes 17 can extend in different transverse directions x, y to reach the respective side surfaces 19 of the substrate 15.

[0085] FIG3 b shows another embodiment of a microphone stack 3 that includes different vent holes 18. In this embodiment, each membrane 14 is pierced at its center by a vent hole 18 that connects the rear volume 17 to the environment above each membrane 14. It should be noted that if a spacer 9 and an elastic layer 8 (not shown) are present, the elastic layer 8 must also include a vent hole 18. In this case, the rear volume 17 is connected to the environment above the elastic layer 8 via a front volume 28 (not shown) formed by the spacer 9.

[0086] In either case, the vent 18 has a smaller diameter than the diameter dm of the membrane 14. This ensures a high acoustic resistance of the vent 18. The acoustic resistance of the vent 18 and the acoustic compliance of the rear volume 17 together form an acoustic high-pass filter. To achieve a low high-pass cutoff frequency, the acoustic resistance of the vent 18 must be large. The vent 18 provides pressure equalization between the rear volume 17 and the environment.

[0087] FIG4 a shows another embodiment of an apparatus 34 for sound detection, sound localization, and beamforming. FIG4 a differs from FIG1 a in that it shows four microphone stacks 3 within display 1. Microphone stacks 3 are spaced apart from one another by a spacing s. Each microphone stack 3 is spaced apart from adjacent microphone stacks 3 in the lateral directions x and y by at least 1 cm and at most 10 cm. However, in the particular embodiment shown in FIG4 a , spacing s can be approximately 7 cm.

[0088] FIG4 b shows an embodiment of a transducer array 5 that can be used in the exemplary embodiment of the device 34 according to FIG4 a. In this example, the transducer array 5 has a square shape in a top view. The length l and width w of the transducer array 5 are at least 1 mm and at most 10 mm. However, in this particular embodiment, the length l and width w of the transducer array 5 can be approximately 2 mm. The transducer array 5 includes four membranes 14. The diameter dm of the membranes 14, which have a circular shape in a top view, is at least 0.1 mm and at most 1.5 mm. In this particular embodiment, the diameter dm can be approximately 0.5 mm. The spacing pm of the membranes 14 (i.e., the distance between their centers) is at least 0.4 mm and at most 3 mm. In this particular embodiment, the spacing pm can be approximately 1.5 mm. The membranes 14 can comprise silicon nitride (SiN) and can have a thickness suitable for the desired application. For example, the thickness can be at least 100 nm and at most 500 nm.

[0089] 4a and 4b, a device 34 for sound detection, sound localization, and beamforming can be realized that has suitable acoustic performance. Although a single transducer array 5 does not cover the entire audio frequency range in terms of directivity and spatial resolution, combining four of them can provide good directivity even at low frequencies. By means of algorithms for sound localization and beamforming, a single transducer array 5 can cover a higher frequency range, while a combination of four transducer arrays can cover a lower frequency range to obtain good directivity. The time delay and spatial resolution of such a device 34 can be determined by directly evaluating the autocorrelation function or by evaluating the Hilbert transform of the autocorrelation function. It can be seen that such a device 34 exhibits a time delay of a few microseconds and can resolve the direction of sound to approximately 2°.

[0090] Another embodiment of a transducer array 5 that can be used in the device 34 according to FIG. 4a is shown in FIG. 4c . In this particular embodiment, the transducer array 5 also has a square shape in top view, while the length l and width w can have the same dimensions as in FIG. 4b , namely, approximately 2 mm. However, in this case, the transducer array 5 includes sixteen membranes 14 arranged in a regular 4×4 grid. Therefore, the diameter dm of each membrane 14 must be smaller, namely, can be approximately 0.1 mm. The pitch pm is approximately 0.4 mm.

[0091] Such a small membrane diameter dm can be achieved using 2D materials. This is because the membranes 14 comprising a layer of 2D material can still achieve acceptable acoustic performance even if they exhibit a small diameter dm. The acoustic performance of the transducer array 5 according to FIG4c is similar to that of the transducer array 5 according to FIG4b. However, due to the increased number of membranes 14, each transducer array 5 can have an increased SNR.

[0092] FIG5 a shows another embodiment of an apparatus 34 for sound detection, sound localization, and beamforming. FIG5 a differs from FIG1 a in that it shows six microphone stacks 3 within display 1. Microphone stacks 3 are spaced apart from one another in lateral direction y by spacings sy. In the specific example shown in FIG5 a , spacings sy can be approximately 3 cm. As shown in FIG4 a , spacings sx in lateral direction x can be approximately 7 cm.

[0093] FIG5 b shows an exemplary embodiment of a transducer array 5 that can be used in an embodiment of the device 34 according to FIG5 a. In this example as well, the transducer array 5 has a square shape in top view with a length l and a width w of approximately 2 mm. The diameter dm of the membrane 14, which has a circular shape in top view, can be approximately 0.1 mm. The pitch pm of the membranes 14, i.e., the distance between their centers, can be approximately 1.7 mm. As in the embodiment according to FIG4 c, the membrane 14 can comprise a layer of 2D material of suitable thickness, for example, only a few nanometers.

[0094] Due to the small diameter dm, the membranes 14 can be arranged relatively far apart even in a transducer array 5 having a small aperture (length l×width w). Even at relatively low frequencies, for example down to 10 kHz, the large distance between the membranes 14 within a single transducer array 5 gives acceptable directivity. It can be seen that the device 34 implemented in Figures 5a and 5b exhibits a spatial resolution of approximately 5°.

[0095] It should be noted again that, in contrast to the case with standard MEMS materials, the membrane 14 comprising a layer of 2D material can be used without the spacer 9 and the additional elastic layer 8. The membrane 14 comprising a layer of 2D material can be mounted directly into the display 1 .

[0096] exist Figure 6 , a schematic diagram of an exemplary embodiment of an electronic device 20 including an apparatus 34 for sound detection, sound localization, and beamforming is shown. The electronic device 20 can be, for example, a smartphone, a smart speaker, a smart TV, a smart TV remote control, or any other household appliance or any other remote control.

[0097] Figures 7a to 7f illustrate an exemplary embodiment of a method for manufacturing a microphone stack 3. Microphone stack 3 includes a transducer device comprising conventional MEMS materials, requiring membrane 14 to be protected by an additional gasket 9 and elastic layer 8. Furthermore, the manufacturing of only one membrane 14 is shown by way of example. However, several membranes are typically provided within a transducer array 5. Because different manufacturing processes can also be applied, the illustrated manufacturing process must be understood as merely an example.

[0098] The method includes providing a substrate 15, as shown in FIG7a. The substrate has a rear surface 21 and a main surface 22. The substrate 15 can comprise Si. A thin film 23 is formed on the main surface 22 of the substrate 15. For example, the thin film 23 can be formed by deposition and subsequent etching to structure the thin film 23. The thin film 23 can comprise SiN.

[0099] In the next step according to FIG. 7 b , grooves 24 are formed in the substrate 15 from the rear surface 21 toward the corresponding portion of the membrane 23. This means that the grooves 24 penetrate the substrate 15 in the region where the membrane 23 is present at the main surface 22 of the substrate 15. The grooves 24 can have a smaller lateral extension than the membrane 23, so that a portion of the membrane 23 is still supported by the substrate 15. The exposed portion of the membrane 23 forms the membrane 14 that can deflect. The grooves 24 form a rear volume 17 beneath each membrane 14. The entire structure forms the transducer array 5.

[0100] An optical pickup device 6 is provided ( FIG. 7 c ). It may include another substrate 16 , such as a semiconductor substrate comprising Si. On the other substrate 16 , at least one light source 25 and at least one photodetector 26 are disposed at a top surface 27 or within the other substrate 16 . For example, the photodetector 26 can be a photodiode or a photodiode array fabricated using a conventional complementary metal oxide semiconductor (CMOS) process. FIG. 7 c illustrates a one-to-one relationship between the membranes 14 and the light sources 25 . However, it is also possible to use only one light source 25 , such as a VCSEL, for a complete microphone stack 3 . In this case, an optical waveguide can be used to provide light to each membrane 14 . The optical pickup device 6 is mounted below the transducer array 5 in the transverse direction z. This means that the rear surface 21 of the substrate 15 is attached to the top surface 27 of the other substrate 16 . Attachment can be achieved by bonding or adhesive. The substrate 15 and the other substrate 16 are aligned so that the optical pickup device 6 can detect the displacement of each membrane 14 using the light source 25 and photodetector 26 passing through the rear volume 17 .

[0101] As shown in FIG7 d , a gasket 9 and an elastic layer 8 are provided. The gasket 9 has a plurality of through-holes 13 extending therethrough. This means that the through-holes 13 connect the first surface 11 and the second surface 12 of the gasket 9. The diameter dg of the through-holes 13 can be greater than the diameter dm of the membrane 14. The gasket 9 is mounted below the elastic layer 8 along the transverse direction z such that the second surface 12 of the gasket 9 is attached to the elastic layer 8.

[0102] The transducer array 5 and the optical pickup device 6 are mounted below the gasket 9 and the elastic layer 8 in the transverse direction z so that the first surface 11 of the gasket 9 is attached to the main surface 22 of the substrate 15. Thus, a front volume 28 is formed above the membrane 14 and below the elastic layer 8. The gasket 9 seals the front volume 28. The membrane 14 is aligned with the through hole 13 of the gasket 9. This means that in the transverse directions x, y, the membrane 14 is arranged in the through hole 13. The thickness t of the gasket 9 can be smaller than the height h of the rear volume 17. The thickness t of the gasket 9 affects the size of the front volume 28 and therefore its resonant frequency. In order to obtain a resonant frequency that is as high as possible, the gasket 9 should be very thin. The final structure of Figure 7d forms the microphone stack 3.

[0103] FIG7e shows a cross-section of two adjacent membrane structures of a microphone stack 3. FIG7e also shows an elastic connector 7 formed around the microphone stack 3 in the transverse directions x and y. The surrounding elastic connector 7 can be formed by molding. The elastic connector 7 can comprise, for example, PDMS.

[0104] In the next step ( FIG. 7 f ), microphone stack 3 is mounted in display 1 together with elastic connector 7 , wherein display 1 surrounds microphone stack 3 at a distance in the lateral directions x, y, and wherein elastic connector 7 mechanically connects display 1 to microphone stack 3 . Elastic connector 7 reduces structure-borne sound from display 1 to transducer array 5 . In FIG. 7 f , vent 18 is not shown. However, it should be noted that vent 18 may also be present according to FIG. 3 a or FIG. 3 b .

[0105] Figures 8a to 8f show an exemplary embodiment of another method for manufacturing a transducer array 5. The transducer array 5 comprises a 2D material and therefore does not require the membrane 14 to be protected by an additional spacer 9 and an elastic layer 8. In addition, Figures 8a to 8e exemplarily illustrate the manufacture of only one membrane. However, as shown in Figure 8f, several membranes 14 are typically provided within the transducer array 5. This method should be considered as an example for manufacturing a membrane 14 comprising a layer of 2D material (in this case, graphene). However, different 2D materials can also be used, such as molybdenum disulfide (MoS2), etc., and their manufacturing processes may be slightly different.

[0106] In FIG8 a , a substrate 15 is shown with a thin film stack on top of it. In this case, the substrate 15 comprises silicon. The thin film stack can be deposited on the substrate 15. The first layer of the thin film stack, which is in mechanical contact with the main surface of the substrate, can comprise silicon oxide (SiO 2 ), thereby forming a silicon oxide layer 29. However, a layer comprising SiN can also be used for the first layer of the thin film stack. The first layer can be deposited by oxidation of the substrate (wet or dry oxidation), by PVD (sputtering or evaporation) or CVD (LPCVD, TEOS, plasma or thermal CVS), or by spin coating and baking (spin-on glass). The second layer on top of the first layer can comprise silicon carbide (SiC), thereby forming a silicon carbide layer 30. The second layer can be deposited by PVD (sputtering or evaporation) or by CVD (plasma or thermal CVD or LPCVD). The first layer is disposed between the substrate 15 and the second layer and serves as a diffusion barrier to prevent particles from diffusing from the second layer into the substrate 15. The third layer of the thin film stack is disposed on top of the second layer so that the second layer is located between the first and third layers. The third layer may include nickel (Ni), thereby forming a nickel layer 31. The third layer may be deposited by PVD (sputtering or evaporation).

[0107] FIG8 b shows the substrate 15 and film stack after heat treatment. The heat treatment may include a rapid thermal annealing (RTA) step at approximately 1000° C. At this temperature, solid-state diffusion can occur, resulting in the transformation of the silicon carbide layer 30 and the nickel layer 31. Carbon is dissolved in the nickel by RTA. After cooling, the carbon segregates to the nickel surface, thereby forming a graphene layer 32 as a thin film on the nickel silicide layer 33.

[0108] In the next step ( FIG. 8 c ), a trench 24 is formed in the substrate from the rear surface 21 toward the silicon oxide layer 29, exposing the silicon oxide layer 29. The trench 24 can be formed by etching. In the case of the silicon substrate 15, the etching can be, for example, deep reactive ion etching (DRIE), xenon difluoride etching (XeF 2 ), low temperature silicon etching, or dry etching using sulfur hexafluoride (SF 6 ) and octafluorocyclobutane (C 4 F 8 ). However, etching can also be performed in a wet chemical manner using, for example, KOH or TMAH.

[0109] 8 d shows that the trench 24 is extended into the silicon oxide layer 29. This means that the silicon oxide layer 29 is removed in the region where the trench 24 is formed in the substrate 15. The silicon oxide layer 29 can be removed by dry oxide etching using fluoroform (CHF3) and oxygen (O2), using tetrafluoromethane (CF4) and argon (Ar), or wet etching using an acid-based reagent (buffered HF (BHF) or pure aqueous HF).

[0110] In the next step according to FIG8 e , the trench 24 is further extended by locally removing the nickel silicide layer 33. This means that the nickel silicide layer 33 is removed in the areas where the trenches are formed in the substrate 15 and the silicon oxide layer 29. The nickel silicide layer 33 can be removed by a nitric acid-based solution or HF. The etching can be isotropic and release the graphene layer 32, which forms the membrane 14 where the trenches 24 are formed below. A drying step may be required after etching. Nitrogen drying, compressed air drying or critical point drying can be used.

[0111] FIG8 f shows a perspective view of the final structure of the transducer array 5. In this embodiment, the transducer array 5 comprises 2×3=6 membranes 14. However, other geometries are also conceivable, such as arrays comprising 2×2=4 or 3×3=9 membranes. The membranes 14 are formed by corresponding portions of the graphene layer 32. The corresponding rear volume 17 is formed by the trenches 24 in the substrate 15, the silicon oxide layer 29, and the nickel silicide layer 33 beneath each membrane 14. In a subsequent processing step, according to FIG7 c, the optical pickup device 6 must be mounted below the transducer array 5 to complete the microphone stack 3. The microphone stack 3 is then inserted into the display 1 using the elastic connector 7, as shown in FIG7 e and 7 f.

[0112] Reference numerals

[0113] 1 display

[0114] 2 Top surface of the monitor

[0115] 3-mic stack

[0116] 4 Other top surface of the microphone stack

[0117] 5 Transducer array

[0118] 6 Optical reading device

[0119] 7 Elastic connector

[0120] 8 Elastic Layer

[0121] 9 gasket

[0122] 10 printed circuit boards

[0123] 11 First surface of gasket

[0124] 12 Second surface of gasket

[0125] 13 Gasket through hole

[0126] 14 Transducer array membrane

[0127] 15. Transducer array substrate

[0128] 16 Another substrate of the optical pickup device

[0129] 17 Transducer rear volume

[0130] 18 vents

[0131] 19 Side surface of substrate

[0132] 20 Electronic devices

[0133] 21 rear surface of substrate

[0134] 22 Main surface of substrate

[0135] 23 films

[0136] 24 grooves

[0137] 25 Light Source

[0138] 26 Photodetector

[0139] 27 Top surface of another substrate

[0140] 28 front volume

[0141] 29 silicon oxide layer

[0142] 30 Silicon carbide layer

[0143] 31 Nickel layer

[0144] 32 graphene layers

[0145] 33 Nickel silicide layer

[0146] 34 devices

[0147] x, y horizontal direction

[0148] z transverse direction

[0149] l Length of the transducer array

[0150] w is the width of the transducer array

[0151] t Thickness of gasket

[0152] dg diameter of the through hole

[0153] pg Pitch of through holes

[0154] dm Diameter of membrane

[0155] h Height of the rear volume

[0156] s Spacing between microphone stacks

[0157] sx The spacing of the microphone stack in the x direction

[0158] sy The spacing of microphone stack in the y direction

[0159] pm membrane spacing

Claims

1. An apparatus (34) for sound detection, sound localization, and beamforming, the apparatus comprising: - a display (1) having a main extension plane, a plurality of microphone stacks (3), wherein the display (1) surrounds each microphone stack (3) in a transverse direction (x, y), the transverse direction extending parallel to a main extension plane of the display (1), and - a plurality of elastic connectors (7), wherein each elastic connector surrounds a respective microphone stack (3) in a transverse direction (x, y) and mechanically connects the respective microphone stack (3) to the display (1), Each microphone stack (3) comprises: - a microelectromechanical transducer array (5), said transducer array (5) comprising a plurality of membranes (14) and corresponding integrated rear volumes (17), said rear volumes (17) being arranged below said membranes (14) in a transverse direction (z), said transverse direction extending perpendicularly to a main extension plane of said display (1), wherein said transducer array further comprises a substrate (15), membranes (23) on a main surface (22) of said substrate (15) and grooves (24) formed from a rear surface (21) of said substrate (15) towards said membranes (23), wherein said membranes (14) are formed by respective portions of the exposed membranes (23) and said rear volumes (17) are formed by said grooves (24), - an optical reading device (6) configured to detect the displacement of each membrane (14) individually, - an elastic layer (8), and - a gasket (9) comprising a plurality of through holes (13) passing through said gasket (9), Wherein, in the transverse direction (z), the gasket (9) is arranged between the elastic layer (8) and the transducer array (5), and the membrane (14) of the transducer array (5) is aligned with the through hole (13) of the gasket (9).

2. The device (34) according to claim 1, wherein Each microphone stack (3) further comprises a 2D material layer as a topmost layer, wherein the membrane (14) of the transducer array (5) is formed by respective portions of the 2D material layer.

3. The device (34) according to claim 1, wherein Each transducer array (5) further comprises a substrate (15) disposed beneath the plurality of membranes (14) in the transverse direction (z), and wherein each rear volume (17) is formed by a cavity in the substrate (15) beneath the respective membrane (14).

4. The device (34) according to claim 3, wherein The transducer array (5) further comprises a vent (18) connecting each rear volume (17) to the environment, the vent (18) being formed by an opening in the substrate (15).

5. The device (34) according to claim 1, wherein The optical reading device (6) comprises at least one light source (25) and at least one photodetector (26), wherein the light source (25) and the photodetector (26) are arranged on a further substrate (16) which is attached to the transducer array (5) below the transducer array (5) in a transverse direction (z).

6. The device (34) according to claim 1, wherein each transducer array (5) has a rectangular shape in a top view, wherein In each transverse direction (x, y), the transducer array (5) has a length (l) and a width (w) of at least 1 mm and at most 10 mm.

7. The device (34) according to claim 1, wherein In the lateral directions (x, y), each transducer array (5) is spaced apart from an adjacent transducer array (5) by at least 1 cm and at most 10 cm.

8. The device (34) according to claim 1, wherein Each membrane (14) within the transducer array (5) has a circular shape in a top view, wherein a diameter (dm) of the circular shape is at least 0.1 mm and at most 1.5 mm.

9. The device (34) according to claim 1, wherein Each membrane (14) within the transducer array (5) is spaced apart from an adjacent membrane (14) by a predetermined pitch (pm), the pitch (pm) being defined as the distance between the centers of the respective membranes, wherein the pitch (pm) is at least 0.4 mm and at most 3 mm.

10. An electronic device (20) comprising the apparatus (34) according to any one of claims 1 to 9.

11. The electronic device (20) according to claim 10, wherein The electronic device (20) is a smart phone, a smart speaker, a smart TV, a smart TV remote control or any other home appliance or any other remote control.

12. A method of manufacturing a device (34) for sound detection, sound localization, and beamforming, the method comprising: - providing a display (1) having a main extension plane, - providing an optical reading device (6), - manufacturing a transducer array (5) by providing a substrate (15), forming a thin film (23) on a main surface (22) of the substrate (15), forming a groove (24) in the substrate (15) from a rear surface (21) of the substrate (15) towards the thin film (23), wherein a plurality of membranes (14) are formed by respective portions of the exposed thin film (23) and corresponding integrated rear volumes (17) are formed by the groove (24), - mounting the optical pickup device (6) below the transducer array (5) along a transverse direction (z) extending perpendicularly to a main extension plane of the display (1) to form a microphone stack (3), wherein manufacturing the microphone stack (3) further comprises: - providing an elastic layer (8), - providing a gasket (9) and mounting said gasket (9) below said elastic layer (8) in a transverse direction (z), said gasket having a plurality of through holes (13) passing through said gasket, and - mounting the transducer array (5) together with the optical pickup (6) below the spacer (9) and the elastic layer (8) in the transverse direction (z), wherein the membrane (14) of the transducer array (5) is aligned with the through-holes (13) of the spacer (9), - forming an elastic connection (7) around the microphone stack (3), the elastic connection (7) surrounding the microphone stack (3) in a transverse direction (x, y) extending parallel to a main extension plane of the display (1), and - Mounting the microphone stack (3) together with the elastic connector (7) in the display (1), wherein the display (1) surrounds the microphone stack (3) at a distance, and wherein the elastic connector (7) mechanically connects the display (1) to the microphone stack (3).

13. The method according to claim 12, wherein: The thin film (23) comprises a layer of 2D material, and the membrane (14) of the transducer array (5) is formed from respective portions of the layer of 2D material.

Images