A MEMS loudspeaker and a hearing aid

By using a frequency-crossing driven MEMS speaker architecture and a miniature acoustic cavity design, the high-frequency response bottleneck of hearing aids is solved, achieving high-fidelity reproduction of high and very high frequencies and sufficient amplitude and smooth response in the mid and low frequency ranges. This improves speech clarity and the naturalness of sound, while reducing the size and weight of hearing aids.

CN123123780APending Publication Date: 2026-11-17HUBEI JIUFENGSHAN LAB
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611535223.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-30
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing hearing aid receivers have bottlenecks in high-frequency response, making it difficult to achieve high-fidelity, low-distortion reproduction of high-frequency and very-high-frequency components. At the same time, the mid-to-low frequency response is not sufficient and smooth enough, affecting speech clarity and the recognition of high-frequency consonants.

Method used

The MEMS speaker architecture employs a frequency-division driven design. Low-frequency MEMS transducers and high-frequency MEMS transducers process mid-low-frequency and high-frequency signals respectively. The audio signals are then separated and fed to the corresponding units through an electronic frequency division network. By combining the high-frequency extension advantage of MEMS devices and the design of miniature acoustic cavities, each unit can operate in its optimal performance frequency band, avoiding intermodulation distortion.

Benefits of technology

It achieves a wide and flat frequency response from low frequencies to over 20kHz, improving speech intelligibility and compensation for users with high-frequency hearing loss. Sound details and dynamic range are faithfully reproduced, resulting in a natural and transparent sound, while reducing the size and weight of the acoustic output module.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123123780A_ABST
    Figure CN123123780A_ABST
Patent Text Reader

Abstract

The application provides a MEMS loudspeaker and a hearing aid, the MEMS loudspeaker comprises a packaging shell, a MEMS loudspeaker chip assembly and a PCB board, the MEMS loudspeaker chip assembly is arranged in the packaging shell, the MEMS loudspeaker chip assembly comprises at least one low-frequency MEMS transducing unit, at least one high-frequency MEMS transducing unit and an electronic frequency division network, the first-order resonance frequency of the low-frequency MEMS transducing unit is 0.5 kHz to 5 kHz, and the low-frequency MEMS transducing unit is used for playing back a low-frequency component in an audio signal; the first-order resonance frequency of the high-frequency MEMS transducing unit is 5 kHz to 12 kHz, and the high-frequency MEMS transducing unit is used for playing back a high-frequency component in the audio signal. The MEMS loudspeaker provided by the application can realize high fidelity and low distortion of high-frequency signals, and meanwhile, the low-frequency band has sufficient amplitude and smooth response frequency characteristics.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of microelectromechanical systems (MEMS) technology, specifically to a MEMS loudspeaker and a hearing aid. Background Technology

[0002] As a crucial hearing compensation device, the fundamental goal of hearing aids is to accurately acquire, process, and reproduce sound to help hearing-impaired users rebuild effective auditory perception. In this process, the receiver, the miniature speaker inside the hearing aid, is the core transducer that ultimately converts the processed electrical signals into audible sound waves. Its performance directly determines the quality, clarity, and naturalness of the final output sound, and profoundly affects the overall size, power consumption, and cost of the device.

[0003] Currently, in-the-ear and behind-the-ear hearing aids commonly use balanced armature or miniature dynamic coil receivers. Balanced armature receivers rely on electromagnetic force to drive a magnetic armature and a linked diaphragm to produce sound. Their advantages include high sensitivity and stable low-to-mid frequency response, but they suffer from significant physical limitations in high-frequency performance. However, the diaphragm and drive system of balanced armature or miniature dynamic coil receivers have a relatively large mass, and the mechanical resonant frequency limits the high-frequency response, typically attenuating significantly after 5kHz to 8kHz. Even with digital algorithms to boost high frequencies, harsh harmonic distortion and "electronic sound" are easily introduced, making it difficult to accurately reproduce natural high-frequency components. Therefore, for users, especially those with high-frequency steep-drop hearing loss, clarity is often unsatisfactory in noisy environments or when listening to high-frequency sound sources. Secondly, the mechanical movement of the magnetic circuit and armature has a delay, resulting in insufficient transient response performance. This makes it inaccurate in reproducing rapidly changing signals such as the onset of percussion instruments in music and plosives in speech, resulting in a slightly muffled sound lacking vibrancy.

[0004] In summary, how to overcome the high-frequency response bottleneck of receivers and achieve high-fidelity, low-distortion reproduction of high-frequency and very-high-frequency components while also having sufficient amplitude and smooth mid-to-low-frequency response, thereby significantly improving speech intelligibility, especially the ability to recognize high-frequency consonants, is a problem that urgently needs to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device and storage medium for segmenting the operating conditions of new energy sources, aiming to solve the problem of how to achieve high-fidelity and low-distortion restoration of high-frequency and very high-frequency components while ensuring that the mid- and low-frequency bands have sufficient amplitude and smooth frequency response characteristics.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: In a first aspect, the present invention provides a MEMS loudspeaker for use in a hearing aid, characterized in that it includes a package housing, a MEMS loudspeaker chip assembly, and a PCB board, wherein an acoustic cavity is formed inside the package housing, and a sound outlet is provided on the package housing to communicate with the acoustic cavity; the PCB board is electrically interconnected with the MEMS loudspeaker chip assembly; the MEMS loudspeaker chip assembly is disposed inside the package housing; and the MEMS loudspeaker chip assembly includes: At least one low-frequency MEMS transducer unit, wherein the first-order resonant frequency of the low-frequency MEMS transducer unit is 0.5kHz to 5kHz, is used to reproduce the mid-to-low frequency components in the audio signal; At least one high-frequency MEMS transducer unit, wherein the first-order resonant frequency of the high-frequency MEMS transducer unit is 5kHz~12kHz, and is used to play back the high-frequency components in the audio signal. And an electronic frequency divider network, used to separate the input audio signal into low-frequency components and high-frequency components, and feed them to the low-frequency MEMS transducer unit and the high-frequency MEMS transducer unit respectively.

[0007] Furthermore, both the low-frequency MEMS transducer and the high-frequency MEMS transducer include a substrate and a diaphragm. A back cavity is formed on the substrate of any transducer. One end of the diaphragm corresponding to the substrate is fixed to the substrate in a first direction, and the other end is suspended above the back cavity to generate out-of-plane bending vibration under the drive of an electrical signal.

[0008] Furthermore, the structural parameters of the low-frequency MEMS transducer unit and the high-frequency MEMS transducer unit are different. The structural parameters of the low-frequency MEMS transducer unit and the high-frequency MEMS transducer unit include at least one of the following: diaphragm thickness, back cavity depth, diaphragm length, diaphragm width, diaphragm clamping method, and number of diaphragms.

[0009] Furthermore, the length of the diaphragm of the low-frequency MEMS transducer unit in the first direction is greater than the length of the diaphragm of the high-frequency MEMS transducer unit in the first direction; and / or, The diaphragm thickness of the low-frequency MEMS transducer is smaller than that of the high-frequency MEMS transducer.

[0010] Furthermore, the low-frequency MEMS transducer unit is a MEMS electromagnetic loudspeaker chip or a MEMS piezoelectric loudspeaker chip, and the high-frequency MEMS transducer unit is a MEMS piezoelectric loudspeaker chip; and / or, Both the low-frequency MEMS transducer and the high-frequency MEMS transducer are MEMS piezoelectric speaker chips. The diaphragm includes, along its thickness direction, a support layer, a bottom electrode layer, a piezoelectric material layer, and a top electrode layer.

[0011] Furthermore, the packaging shell has a first inner wall and a second inner wall disposed opposite to each other in the thickness direction, the low-frequency MEMS transducer unit is disposed on the first inner wall, and the high-frequency MEMS transducer unit is disposed on the second inner wall.

[0012] Furthermore, the diaphragm includes, in sequence along its thickness direction, a support layer, a bottom electrode layer, a piezoelectric material layer, and a top electrode layer.

[0013] Furthermore, the MEMS speaker chip assembly includes at least two low-frequency MEMS transducer units and at least two high-frequency MEMS transducer units. At least two of the low-frequency MEMS transducer units are arranged in a low-frequency subarray, and at least two of the high-frequency MEMS transducer units are arranged in a high-frequency subarray; The low-frequency subarray and the high-frequency subarray together constitute a loudspeaker array.

[0014] Furthermore, the speaker array has a first operating mode and a second operating mode; In the first operating mode, all MEMS transducer units in the low-frequency subarray and the high-frequency subarray are configured to be synchronously driven; In the second operating mode, driving signals with a predetermined phase difference or time delay are applied between each low-frequency MEMS transducer unit in the low-frequency subarray and / or between each high-frequency MEMS transducer unit in the high-frequency subarray to form a directional sound beam in space.

[0015] In a second aspect, the present invention also provides a hearing aid, characterized in that it includes: a functional module and a MEMS speaker provided in the first aspect; The functional module is electrically connected to the MEMS speaker, and the MEMS speaker converts the electrical signal output by the functional module into an acoustic signal.

[0016] Furthermore, the hearing aid also includes a housing; The housing covers the functional module and the MEMS speaker, and the housing also contains a main control SoC chip and a microphone array; The microphone array is disposed on the surface of the housing and is used to pick up external sound wave signals; The main control SoC chip is electrically connected to the microphone array and the MEMS speaker respectively, and is used to perform digital signal processing on the signal picked up by the microphone array and transmit the processed audio signal to the MEMS speaker.

[0017] The MEMS loudspeaker provided by this invention has the following significant technical effects: 1. This invention employs a frequency-division driven MEMS speaker architecture. By using a high-frequency MEMS transducer unit with a first-order resonant frequency of 5kHz~12kHz, specifically designed for high-frequency signal playback, it fully leverages the high-frequency extension advantage of MEMS devices, enhancing consonant discrimination and sound detail. Furthermore, a low-frequency MEMS transducer unit with a first-order resonant frequency of 0.5kHz~5kHz is primarily responsible for outputting low-frequency signals from speech. This transducer unit has sufficient amplitude and a smooth response, ensuring adequate sound pressure in the speech energy concentration frequency band. An electronic frequency division network divides the full-band audio signal into multiple frequency bands according to a preset division point. Each frequency band is output to its corresponding MEMS transducer subunit for driving. Each transducer subunit operates only in its optimal performance frequency band, avoiding intermodulation distortion and nonlinear distortion caused by full-band operation, resulting in higher output sound pressure level, a flatter frequency response, and lower distortion, among other comprehensive acoustic performance characteristics. Therefore, the MEMS speaker provided by this invention can achieve high-fidelity, low-distortion reproduction of high-frequency signals while providing sufficient amplitude and a smooth frequency response in the mid-to-low frequency range.

[0018] 2. By using the aforementioned MEMS speaker as the receiver, the hearing aid achieves a wide and flat frequency response from low frequencies to over 20kHz. This allows for lossless amplification of high-frequency components containing rich speech information, solving the problem of insufficient high frequencies in traditional receivers and greatly improving speech clarity, especially for users with high-frequency hearing loss. Its microsecond-level transient response faithfully reproduces the details, speed, and dynamic range of sound, resulting in a more natural and transparent sound, significantly improving the listening experience in scenarios such as music appreciation and conversation.

[0019] 3. The miniaturization of the MEMS chip itself, combined with the miniature acoustic rear cavity, reduces the weight of the entire acoustic output module by 30% to 50% compared to traditional balanced armature receivers. This provides core hardware support for developing next-generation hearing aids that are smaller, fit the ear canal anatomy better, and are more discreet and comfortable to wear. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a transducer unit provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a MEMS speaker chip assembly provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a MEMS loudspeaker provided in another embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of a MEMS speaker chip assembly provided in another embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a MEMS loudspeaker provided in another embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of a hearing aid provided in an embodiment of the present invention.

[0021] Explanation of icon numbers: 100. MEMS speaker; 1. Encapsulation shell; 11. Sound outlet; 2. MEMS speaker chip assembly; 3. PCB board; 31. Pressure relief hole; 4. Low-frequency MEMS transducer unit; 5. High-frequency MEMS transducer unit; 6. Substrate; 7. Diaphragm; 8. Back cavity; 200. Hearing aid; 201. Housing; 202. Earplug; 203. Cannula; 204. Interaction component; 205. Battery compartment. Detailed Implementation

[0022] The implementation methods of this solution will be described in further detail below. Obviously, the described embodiments are only a part of the embodiments of this solution, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this solution can be combined with each other.

[0023] Currently, in-ear and behind-the-ear hearing aids commonly use balanced armature or miniature dynamic coil receivers. Balanced armature receivers rely on electromagnetic force to drive a magnetic armature and a linked diaphragm to produce sound. Their advantages include high sensitivity and stable low-to-mid frequency response, but they suffer from significant physical limitations in high-frequency performance. However, the diaphragm and drive system of balanced armature or miniature dynamic coil receivers have a relatively large mass, and the mechanical resonant frequency limits the high-frequency response, typically attenuating significantly after 5kHz to 8kHz. Even with digital algorithms to boost high frequencies, harsh harmonic distortion and "electronic sound" are easily introduced, making it difficult to accurately reproduce natural high-frequency components. Therefore, for users, especially those with high-frequency steep-drop hearing loss, clarity is often unsatisfactory in noisy environments or when listening to high-frequency sound sources. Secondly, the mechanical movement of the magnetic circuit and armature has a delay, resulting in insufficient transient response performance. This makes it inaccurate in reproducing rapidly changing signals such as the onset of percussion instruments in music and plosives in speech, resulting in a slightly muffled sound lacking vibrancy.

[0024] When conventional MEMS loudspeakers are directly transplanted into hearing aids, high-frequency extension can be achieved solely by the diaphragm itself, but low-frequency sound pressure is insufficient, resulting in an imbalance of sound pressure across the entire frequency range, which cannot be adapted to the load conditions of the hearing aid's ear canal.

[0025] In summary, how to overcome the high-frequency response bottleneck of receivers and achieve high-fidelity, low-distortion reproduction of high-frequency and very-high-frequency components while also having sufficient amplitude and smooth mid-to-low-frequency response, thereby significantly improving speech intelligibility, especially the ability to recognize high-frequency consonants, is a problem that urgently needs to be solved in this field.

[0026] Based on this, such as Figure 1 As shown, this embodiment of the invention provides a MEMS speaker 100 for use in a hearing aid 200. The MEMS speaker 100 includes a package shell 1, a MEMS speaker chip assembly 2, and a PCB board 3. An acoustic cavity is formed inside the package shell 1, and a sound outlet 11 communicating with the acoustic cavity is provided on the package shell 1. The PCB board 3 is electrically interconnected with the MEMS speaker chip assembly 2. The MEMS speaker chip assembly 2 is disposed inside the package shell 1. The MEMS speaker chip assembly 2 includes at least one low-frequency MEMS transducer 4, at least one high-frequency MEMS transducer 5, and an electronic frequency divider network. The first-order resonant frequency of the low-frequency MEMS transducer 4 is 0.5kHz to 5kHz, and it is used to reproduce the mid-low frequency components in the audio signal. The first-order resonant frequency of the high-frequency MEMS transducer 5 is 5kHz to 12kHz, and it is used to reproduce the high-frequency components in the audio signal. The electronic frequency divider network is used to separate the input audio signal into low-frequency components and high-frequency components, and feed them to the low-frequency MEMS transducer 4 and the high-frequency MEMS transducer 5, respectively.

[0027] This invention employs a frequency-division driven MEMS loudspeaker 100. A high-frequency MEMS transducer 5 with a first-order resonant frequency of 5kHz to 12kHz is specifically designed for high-frequency signal playback, fully leveraging the high-frequency extension advantage of MEMS devices to enhance consonant discrimination and sound detail. Furthermore, a low-frequency MEMS transducer 4 with a first-order resonant frequency of 0.5kHz to 5kHz primarily outputs low-frequency signals from speech, providing sufficient amplitude and smooth response to ensure adequate sound pressure in the speech energy concentration frequency band. An electronic frequency division network divides the full-band audio signal into multiple frequency bands according to preset division points, outputting each band to its corresponding MEMS transducer subunit for driving. Each transducer subunit operates only in its optimal performance frequency band, avoiding intermodulation distortion and nonlinear distortion caused by full-frequency operation, resulting in higher output sound pressure levels, a flatter frequency response, and lower distortion, among other comprehensive acoustic performance characteristics. Therefore, the MEMS speaker 100 provided by the present invention can achieve high-fidelity and low-distortion reproduction of high-frequency signals, while also providing sufficient amplitude and smooth frequency response characteristics in the mid-to-low frequency range.

[0028] The acoustic cavity can be naturally formed by the encapsulation structure including the outer shell 1, or it can be specially prepared through micromachining or precision injection molding. The acoustic cavity provides suitable acoustic compliance load and extends the low-frequency response. A micro acoustic damping structure can also be added to the acoustic cavity to suppress the resonance peak amplitude and smooth the frequency response curve. A micro acoustic conduit or sound outlet structure can also be integrated into the sound wave output path to complete sound wave coupling matching and optimize the acoustic impedance entering the ear canal.

[0029] See also Figure 2 and Figure 3 Furthermore, both the low-frequency MEMS transducer 4 and the high-frequency MEMS transducer 5 include a substrate 6 and a diaphragm 7. A back cavity 8 is formed on the substrate 6 of any transducer. One end of the diaphragm 7 corresponding to the substrate 6 is fixed to the substrate 6 in a first direction, and the other end is suspended above the back cavity 8 to generate out-of-plane bending vibration under the drive of an electrical signal.

[0030] It should be noted that the transducer unit can be a core electroacoustic transducer element fabricated using microelectromechanical systems (MEMS) technology. Its structure employs semiconductor microfabrication processes (surface silicon, bulk silicon, etc.) to form a micro-diaphragm on single-crystal silicon or other suitable substrates. The diaphragm can be driven by piezoelectric, electrostatic, electromagnetic, or other methods. Thanks to the silicon-based material and micrometer-scale structural design, the diaphragm's mass is only in the microgram range, possessing a very high stiffness-to-mass ratio. This results in a mechanical resonant frequency higher than traditional dynamic and balanced armature loudspeakers, providing a physical basis for achieving a flat, wideband acoustic response from tens of hertz to over 20 kHz. Simultaneously, the diaphragm has a small moment of inertia, exhibiting a microsecond-level mechanical response speed and excellent transient acoustic characteristics.

[0031] In one embodiment, the low-frequency MEMS transducer 4 and the high-frequency MEMS transducer 5 can be fabricated simultaneously on the same silicon wafer and share the same substrate 6, or they can be fabricated on different silicon wafers.

[0032] The low-frequency MEMS transducer and the high-frequency MEMS transducer have different structural parameters, so that the low-frequency MEMS transducer is used to reproduce the low-to-mid-frequency components of the audio signal, and the high-frequency MEMS transducer is used to reproduce the high-frequency components of the audio signal. In one embodiment, frequency response differentiation is achieved by adjusting the structural parameters of the diaphragm 7. The structural parameters of the low-frequency MEMS transducer 4 and the high-frequency MEMS transducer 5 include at least one of the following: diaphragm thickness, cavity depth, diaphragm length, diaphragm width, diaphragm clamping method, and number of diaphragms. Preferably, the low-frequency MEMS transducer 4 adopts a long cantilever beam configuration, or a multi-cantilever beam or widened cantilever beam configuration, to reduce the inherent stiffness of its overall structure, thereby placing its first-order resonant frequency in the range of 0.5kHz to 5kHz. Thus, the low-frequency MEMS transducer 4 is mainly used to output the low-to-mid-frequency components of the speech signal, ensuring that the speech energy concentration frequency band has sufficient sound pressure level. The high-frequency MEMS transducer 5 can improve the equivalent stiffness of its overall structure by shortening the cantilever beam length or changing its shape, thereby setting its first-order resonant frequency within the range of 5kHz to 12kHz. Therefore, the high-frequency MEMS transducer 5 is specifically designed for reproducing high-frequency components in audio signals, fully leveraging the high-frequency extension advantages of MEMS devices to effectively improve the clarity of consonants and the ability to reproduce sound details.

[0033] See Figure 1 and Figure 3 In one embodiment, the length of the diaphragm 7 of the low-frequency MEMS transducer 4 in the first direction is greater than the length of the diaphragm 7 of the high-frequency MEMS transducer 5 in the first direction; see also Figure 4 In another embodiment, the thickness of the diaphragm 7 of the low-frequency MEMS transducer 4 is smaller than the thickness of the diaphragm 7 of the high-frequency MEMS transducer 5.

[0034] In some other embodiments, the low-frequency MEMS transducer unit and the high-frequency MEMS transducer unit have different diaphragm shapes or clamping methods to give the low-frequency MEMS transducer unit a larger equivalent mass and / or a smaller equivalent stiffness, thereby making its first-order resonant frequency lower than that of the high-frequency MEMS transducer unit, but not limited to this.

[0035] In the technical solution of this application, the low-frequency MEMS transducer unit and the high-frequency MEMS transducer unit may selectively and independently adopt piezoelectric driving, electrostatic driving and electromagnetic driving methods, but are not limited to these.

[0036] Furthermore, the low-frequency MEMS transducer 4 is a MEMS electromagnetic speaker chip or a MEMS piezoelectric speaker chip, and the high-frequency MEMS transducer 5 is a MEMS piezoelectric speaker chip.

[0037] When the low-frequency MEMS transducer 4 or the high-frequency MEMS transducer 5 is a MEMS piezoelectric speaker chip, the diaphragm 7 further comprises, along its thickness direction, a support layer, a bottom electrode layer, a piezoelectric material layer, and a top electrode layer. The support layer is made of single-crystal silicon or silicon dioxide, providing mechanical strength and structural support for the diaphragm 7. The bottom electrode layer and the top electrode layer are each independently made of at least one of titanium (Ti), platinum (Pt), and gold (Au). The bottom electrode layer and the top electrode layer are respectively disposed on the upper and lower surfaces of the piezoelectric material layer, used to apply a driving electric field to the piezoelectric layer. The piezoelectric material layer includes at least one of lead zirconate titanate (PZT), aluminum nitride (AlN), and scandium-doped aluminum nitride (AlScN). The diaphragm 7 can be designed in a square, polygonal, or circular shape, with one end fixed to the substrate 6 by an anchoring structure, and the other end and the central region suspended, generating out-of-plane bending vibration under the action of an applied electric field. An electrical signal is applied to the piezoelectric material layer via the top and bottom electrodes. Based on the inverse piezoelectric effect, the piezoelectric material deforms under the excitation of the electric field, causing the entire diaphragm 7 to undergo out-of-plane bending vibration. The periodic vibration of the diaphragm 7 drives the surrounding air medium to form sound waves with alternating density, completing the conversion of the electrical signal into an audible sound signal.

[0038] Furthermore, the MEMS loudspeaker 100 may selectively employ auxiliary films and structures such as flexible coupling layers, mass blocks, and slotted structures. Flexible coupling layers can be disposed on the surface of the diaphragm 7 or sandwiched between layers to reduce the overall stiffness of the diaphragm 7 and improve low-frequency response sensitivity. Mass blocks can be disposed in the central region or free end of the diaphragm 7 to adjust the resonant frequency by increasing the equivalent mass of the diaphragm 7, thereby extending the frequency response to lower frequencies. Slotted structures are formed at the edge region of the diaphragm 7 or at the root of the cantilever to release internal stress in the films and improve device reliability and acoustic consistency. The aforementioned auxiliary films and structures can be selectively combined according to requirements, and this application does not impose any limitations on this.

[0039] It should be noted that the MEMS speaker chip assembly 2 can be assembled on a miniature rigid or flexible PCB board 3 using flip-chip bonding or wire bonding processes. The PCB board 3 is provided with pressure relief holes 31, and PCB traces lead the chip electrodes to external pads. The PCB board 3 can be sealed inside the package housing 1 together with each transducer unit, or the PCB board 3 can be sealed to the package housing 1 to seal each of the transducer units inside the package housing 1. The size of the MEMS speaker 100 can be controlled at the millimeter level, and it can be directly integrated into the earpiece 202 of the RIC hearing aid 200.

[0040] The MEMS loudspeaker 100 may include one or more of the encapsulation housings 1. One encapsulation housing 1 may encapsulate one transducer unit or multiple transducer units.

[0041] In some embodiments, the low-frequency MEMS transducer 4 and the high-frequency MEMS transducer 5 are packaged in a single housing 201 to form a multi-unit integrated MEMS speaker 100 module, which improves performance while maintaining a compact structure and is easy to place within the limited internal space of the hearing aid 200.

[0042] Please see Figure 1 Taking a low-frequency MEMS transducer 4 and a high-frequency MEMS transducer 5 encapsulated within the encapsulation housing 201 as an example: Low-frequency MEMS transducer unit 4 and high-frequency MEMS transducer unit 5 can be fabricated on the same silicon wafer. Different acoustic output characteristics can be obtained by different structural designs on the same silicon wafer to obtain the low-frequency MEMS transducer unit 4 and the high-frequency MEMS transducer unit 5. For example, the low-frequency MEMS transducer unit 4 and the high-frequency MEMS transducer unit 5 adopt the same or similar process flow, and the basic structure of the vertical film layers such as the support layer and piezoelectric layer is consistent. The frequency response can be distinguished by adjusting the size parameters of the diaphragm 7, as explained above, and will not be repeated here. After the wafer is fabricated, it is diced to obtain a single MEMS chip integrating dual transducer subunits. This chip is packaged into the same package shell 1 in the same way as described above.

[0043] The low-frequency MEMS transducer unit 4 and the high-frequency MEMS transducer unit 5 can also be fabricated on different silicon wafers. They can not only obtain different acoustic output characteristics by relying on differentiated structural designs (such as the structural parameters mentioned above), but also obtain different acoustic output characteristics by using different working principles. For example, the low-frequency MEMS transducer unit 4 uses electromagnetic principles and fabricates a cavity, coil and diaphragm 7 structure through micro-nano fabrication, which can output sufficient low-frequency sound pressure. The high-frequency MEMS transducer unit 5 uses piezoelectric principles to fabricate chips, giving full play to the advantages of piezoelectric MEMS in terms of good high-frequency performance and fast transient response.

[0044] The low-frequency MEMS transducer 4 and the high-frequency MEMS transducer 5 share the acoustic cavity and sound outlet 11. After packaging, the whole unit has a millimeter-level size, which is suitable for the installation space of the RIC hearing aid 200 earplug 202.

[0045] In a specific implementation, the low-frequency MEMS transducer unit 4 and the high-frequency MEMS transducer unit 5 can adopt a parallel packaging structure (see reference). Figure 1 Alternatively, a stacked packaging structure can be used. Figure 4This is a schematic diagram of a MEMS speaker 100 with a stacked packaging structure according to an embodiment of this application. MEMS transducer units operating at different frequency bands are stacked along the thickness direction and share the same sound outlet 11. The packaging shell 1 has a first inner wall and a second inner wall disposed opposite each other in the thickness direction. The low-frequency MEMS transducer unit 4 is disposed on the first inner wall, and the high-frequency MEMS transducer unit 5 is disposed on the second inner wall. The packaging shell 1 has a sound outlet 11 communicating with the acoustic cavity. The PCB board 3 is electrically interconnected with the MEMS transducer units. The low-frequency MEMS transducer unit 4 and the high-frequency MEMS transducer unit 5 share the packaging shell 1 and a portion of the acoustic cavity. When using the stacked packaging structure, the packaging shell 1 encapsulates at least two transducer units. The thickness of the MEMS transducer unit itself is typically less than 1.5 mm, and the total thickness after stacking can still be controlled within 3-4 mm, allowing it to be assembled into the earpiece 202 of in-ear hearing aids such as RIC hearing aids. Compared to the parallel packaging structure, the stacked packaging structure has a higher space utilization rate and is more suitable for the small internal space of the earpiece 202 in the hearing aid 200.

[0046] In other implementations, please refer to Figure 6 Two or more transducer units can also be packaged independently and powered by the same substrate to form a multi-unit integrated MEMS speaker module 100. This solution simplifies the circuit while allowing for flexible cavity design, configuring suitable resonant cavities for different sound-producing units, and further improving the acoustic performance of each unit.

[0047] For further details, please refer to Figure 5 The MEMS speaker chip assembly 2 includes at least two low-frequency MEMS transducer units 4 and at least two high-frequency MEMS transducer units 5; the at least two low-frequency MEMS transducer units 4 are arranged as a low-frequency subarray, and the at least two high-frequency MEMS transducer units 5 are arranged as a high-frequency subarray; the low-frequency subarray and the high-frequency subarray together constitute a speaker array.

[0048] Furthermore, the speaker array has a first operating mode and a second operating mode. In the first operating mode, all transducers in the low-frequency subarray and the high-frequency subarray are configured to drive synchronously, which effectively increases the effective radiation area, significantly improves the overall output sound pressure level, and enhances radiation efficiency. In the second operating mode, driving signals with a predetermined phase difference or time delay are applied between the low-frequency MEMS transducers 4 in the low-frequency subarray and / or between the high-frequency MEMS transducers 5 in the high-frequency subarray to form a directional sound beam in space. In the second operating mode, the directional sound beam concentrates sound energy towards the ear canal, improving sound energy transmission efficiency and reducing unwanted sound diffusion. Moreover, the directional sound beam can reduce sound leakage from the ear canal, playing a positive role in suppressing acoustic feedback howling of the hearing aid 200 and maintaining system stability under higher gain conditions.

[0049] The present invention also proposes a hearing aid 200, which includes a functional module and the aforementioned MEMS speaker 100; the functional module is electrically connected to the MEMS speaker 100, and the MEMS speaker 100 converts the electrical signal output by the functional module into an acoustic signal.

[0050] Please see Figure 7 In some embodiments, the hearing aid 200 further includes a housing 201, which covers the functional module. The housing 201 also houses a main control SoC chip and a microphone array. The microphone array is disposed on the surface of the housing 201 and is used to pick up external sound wave signals. The main control SoC chip is electrically connected to the microphone array and the MEMS speaker 100, respectively. The main control SoC chip performs digital signal processing on the signals picked up by the microphone array and transmits the processed audio signals to the MEMS speaker 100. Utilizing the adjustable frequency band and thin design advantages of the MEMS speaker 100, multiple transducer units operating in different frequency bands are combined to construct a full-range sound module, ensuring the complete acoustic performance of the hearing aid 200 from low to high frequencies, improving the user's auditory perception and speech recognition abilities. Furthermore, the combined overall sound unit still meets the spatial size constraints of the hearing aid 200's in-ear end.

[0051] It should be noted that the main control SoC chip primarily performs digital signal processing tasks, including noise suppression, feedback cancellation, frequency band gain adjustment, and beamforming, and is electrically connected to environmental sensors such as accelerometers and gyroscopes. These environmental sensors collect user motion data and can trigger fall detection or scene-adaptive adjustments. The PCB board 3 simultaneously houses and electrically connects the power management circuit, the main control SoC chip, other functional chips, and discrete electronic components. It can also optionally integrate a wireless communication module to support Bluetooth connectivity, data synchronization, and remote debugging. An interactive component 204 can also be installed on the housing 201. This interactive component 204 can be a physical button or a touch sensor, allowing users to perform operations such as volume adjustment, program switching, and answering calls. A battery compartment 205 can also be installed inside the housing 201. The battery compartment 205 can be configured as a rechargeable lithium battery compartment 205 or a replaceable zinc-air battery compartment 205 to adapt to different usage habits and battery life requirements.

[0052] The functional modules include a data acquisition module, a digital signal processing module, and a driving module. The MEMS speaker 100 can be driven directly by the audio signal output from the SoC chip or by the driving module. The driving module preferably uses a high-voltage, low-distortion Class D or Class H audio power amplifier and is optimized for the capacitive load characteristics of the MEMS speaker 100. The driving module receives digital audio streams or analog audio signals from the digital signal processing module and converts them into an analog driving voltage sufficient to drive the MEMS diaphragm 7 to vibrate linearly in a large amplitude. The driving module and the MEMS speaker 100 unit can be co-optimized at the circuit and impedance matching levels to maximize electro-acoustic conversion efficiency and effectively suppress nonlinear distortion, thereby improving the output sound pressure level and sound quality performance of the MEMS speaker 100.

[0053] The digital signal processing module's algorithm is specifically tuned for the physical characteristics of the MEMS speaker 100. Leveraging its high high-frequency gain and low distortion, the algorithm performs gain equalization in the mid-to-high frequency range to suppress distortion introduced by resonance peaks; it also implements gain constraints for higher-order modes and the high-frequency range of ear canal elevation to prevent high-frequency overshoot. Simultaneously, relying on the device's fast transient response characteristics, more robust multi-channel transient noise suppression and transient enhancement algorithms can be enabled, preserving and enhancing transient speech details while suppressing sudden strong noise, thus improving speech intelligibility in complex environments. The electronic frequency division network can be implemented as a software functional unit within the digital signal processing module or as an independent analog or digital hardware circuit.

[0054] Furthermore, the hearing aid 200 also includes an earplug 202 and a conduit 203. The earplug 202 is worn inside the user's ear canal, and its shape conforms to the contour of the ear canal to achieve a secure fit and acoustic seal. It is generally made of silicone or other biocompatible materials. The conduit 203 connects the earplug 202 to the main housing 201. The earplug 202 has an internal receiving cavity with a diameter of 3-4 mm. The MEMS speaker 100 is disposed inside the receiving cavity, converting electrical signals into sound signals and feeding them directly into the ear canal, reducing sound energy transmission loss. The housing 201 is worn behind the user's ear, and the functional module is disposed inside the housing 201. The conduit 203 has a wire inside, which transmits the audio signal and power supply output from the housing 201 to the MEMS speaker 100.

[0055] By using MEMS speaker 100 as the core sound-generating component, the size and weight of the acoustic output module are significantly reduced. This reduction in size and weight can be directly used to achieve a smaller in-ear shell 201, a lightweight behind-the-ear device, or to free up more assembly space for other components such as batteries, ultimately reducing the invasiveness and pressure of wearing the hearing aid 200 and improving long-term wearing comfort.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A MEMS loudspeaker for use in a hearing aid, characterized in that, The device includes a package housing, a MEMS speaker chip assembly, and a PCB board. An acoustic cavity is formed inside the package housing, and a sound outlet is provided on the package housing to communicate with the acoustic cavity. The PCB board is electrically interconnected with the MEMS speaker chip assembly, which is disposed inside the package housing. The MEMS speaker chip assembly includes: At least one low-frequency MEMS transducer unit, wherein the first-order resonant frequency of the low-frequency MEMS transducer unit is 0.5kHz to 5kHz, is used to reproduce the mid-to-low frequency components in the audio signal; At least one high-frequency MEMS transducer unit, wherein the first-order resonant frequency of the high-frequency MEMS transducer unit is 5kHz~12kHz, and is used to play back the high-frequency components in the audio signal. And an electronic frequency divider network, used to separate the input audio signal into low-frequency components and high-frequency components, and feed them to the low-frequency MEMS transducer unit and the high-frequency MEMS transducer unit respectively.

2. The MEMS loudspeaker as described in claim 1, characterized in that, Both the low-frequency MEMS transducer and the high-frequency MEMS transducer include a substrate and a diaphragm. A back cavity is formed on the substrate of any transducer. One end of the diaphragm corresponding to the substrate is fixed to the substrate in a first direction, and the other end is suspended above the back cavity to generate out-of-plane bending vibration under the drive of an electrical signal.

3. The MEMS loudspeaker as described in claim 2, characterized in that, The low-frequency MEMS transducer and the high-frequency MEMS transducer have different structural parameters. The structural parameters of the low-frequency MEMS transducer and the high-frequency MEMS transducer include at least one of the following: diaphragm thickness, back cavity depth, diaphragm length, diaphragm width, diaphragm clamping method, and number of diaphragms.

4. The MEMS loudspeaker as described in claim 3, characterized in that, The length of the diaphragm of the low-frequency MEMS transducer in the first direction is greater than the length of the diaphragm of the high-frequency MEMS transducer in the first direction; and / or, The diaphragm thickness of the low-frequency MEMS transducer is smaller than that of the high-frequency MEMS transducer.

5. The MEMS loudspeaker as described in claim 2, characterized in that, The low-frequency MEMS transducer is a MEMS electromagnetic speaker chip or a MEMS piezoelectric speaker chip; and / or, the high-frequency MEMS transducer is a MEMS piezoelectric speaker chip; Both the low-frequency MEMS transducer and the high-frequency MEMS transducer are MEMS piezoelectric speaker chips. The diaphragm includes, along its thickness direction, a support layer, a bottom electrode layer, a piezoelectric material layer, and a top electrode layer.

6. The MEMS loudspeaker as claimed in claim 1, characterized in that, The encapsulation shell has a first inner wall and a second inner wall disposed opposite to each other in the thickness direction. The low-frequency MEMS transducer unit is disposed on the first inner wall, and the high-frequency MEMS transducer unit is disposed on the second inner wall.

7. The MEMS loudspeaker as claimed in claim 1, characterized in that, The MEMS speaker chip assembly includes at least two low-frequency MEMS transducer units and at least two high-frequency MEMS transducer units. At least two of the low-frequency MEMS transducer units are arranged in a low-frequency subarray, and at least two of the high-frequency MEMS transducer units are arranged in a high-frequency subarray; The low-frequency subarray and the high-frequency subarray together constitute a loudspeaker array.

8. The MEMS loudspeaker as described in claim 7, characterized in that, The speaker array has a first operating mode and a second operating mode; In the first operating mode, all MEMS transducer units in the low-frequency subarray and the high-frequency subarray are configured to be synchronously driven; In the second operating mode, driving signals with a predetermined phase difference or time delay are applied between each low-frequency MEMS transducer unit in the low-frequency subarray and / or between each high-frequency MEMS transducer unit in the high-frequency subarray to form a directional sound beam in space.

9. A hearing aid, characterized in that, Includes functional modules and the MEMS speaker as described in any one of claims 1 to 8; The functional module is electrically connected to the MEMS speaker, and the MEMS speaker converts the electrical signal output by the functional module into an acoustic signal.

10. The hearing aid as described in claim 9, characterized in that, The hearing aid also includes a housing; The housing covers the functional module and the MEMS speaker, and the housing also contains a main control SoC chip and a microphone array; The microphone array is disposed on the surface of the housing and is used to pick up external sound wave signals; The main control SoC chip is electrically connected to the microphone array and the MEMS speaker respectively, and is used to perform digital signal processing on the signal picked up by the microphone array and transmit the processed audio signal to the MEMS speaker.