Sound quality in technology has never been more important than today, particularly as more sophisticated smartphones and a growing number of smart devices are now able to process language and voice commands. Ideally, the components responsible for sound sensing in modern devices should also deliver high-quality audio for a more pleasant voice-call experience, come in a small form factor for easy integration into already portable gear, and deliver good performance amid noisy background environments without consuming too much power.
Little wonder, then, that prospects will be booming in the years to come for MEMS, or microelectromechanical systems, microphones—a class of miniscule devices using acoustic sensors in combination with a pressure-sensitive diaphragm and an application-specific integrated circuit (ASIC), fabricated on silicon wafers, to sense voice and sound.
Global revenue for MEMS microphones reached $1.1 billion in 2017, up 10% from $987.0 million a year earlier. By 2021, IHS Markit forecasts show revenue worldwide will climb to a solid $1.5 billion, equivalent to some 6.6 billion units in shipments, compared to 3.9 billion units in 2016.
The MEMS microphones market is driven primarily by mobile handsets, with microphone consumption per smartphone on the rise since 2010.
Aside from smartphones, MEMS microphones derive significant revenue from their use in earphones, laptops, hearing aids, and smart speakers. The latter, in fact, is the fastest-growing application for MEMS microphones: smart speakers are projected to enjoy a five-year compound annual growth rate of 36% from 2016 to 2021, IHS Markit data show.
Smart speakers are part of the MEMS microphone segment catering to Internet of Things (IoT) devices. Other IoT devices in the segment using MEMS microphones include wearables, remote-control gear, and home appliances like refrigerators, air conditioners, and service robots.
Features and technology in MEMS microphones
Compared to traditional electret condenser microphones, MEMS microphones carry several advantages, including lower power consumption and higher signal-to-noise ratio (SNR). A higher SNR value translates into better capability for the MEMS microphone to distinguish a sound from background noise, especially in far-field applications. In the end-product (e.g., handset) this means better call- and voice-recording quality, as well as more efficient voice-recognition functionality in digital assistants like Apple’s Siri, Google’s Assistant, and Amazon’s Alexa.
At present, capacitive technology is the best choice for high-SNR microphones, without compromising the size and the form factor of the device. High-SNR microphones feature SNR levels higher than, or equal to, 64 decibels. In the past, low-SNR microphones, featuring SNR levels of less than 60 decibels, were the standard device in many handsets. However, limitations in their performance became clear in situations like voice recordings and voice commands, where there was some distance between the microphone and the source of the sound being recorded.
A second technology for MEMS microphones is piezo, first introduced in small-form-factor MEMS microphones in mid-2016, featuring SNR levels of 62-63 decibels. Piezo MEMS microphones offer ultra-low-power capabilities and can support rugged features, such as waterproofing in smartphones. A higher SNR version of 64-66 decibels was introduced at the end of 2016, but proved too expensive and too big for use in handsets. The target market for piezo is in consumer-oriented applications that can accept large-form-factor MEMS microphones while still benefiting from low-power consumption. Examples include portable smart speakers and remote-control devices.
Most MEMS microphones today are of the analog type, like those used in all Apple iPhones. Analog MEMS microphones account for nearly 80% of the total global market at present. In handsets, Samsung is the only major player using digital MEMS microphones—two in each of their flagship smartphones. More expensive than their analog counterparts, digital MEMS microphones are less sensitive to electromagnetic interference. When used in laptops, for example, they offer better immunity to interference caused by the laptop’s large liquid-crystal display (LCD).
Suppliers and buyers
The top three suppliers of MEMS microphones based on revenue are Knowles of Illinois and the Chinese makers Goertek and AAC Technologies. Together, the three represent nearly 80% of global MEMS microphones revenue. In shipments, the top three are Infineon of Germany and the Japanese manufacturers Omron and NJRC.
The foremost buyer of MEMS microphones is Apple, accounting for 31% of total units shipped in 2017. Apple deploys four MEMS microphones in each iPhone today, up from two in 2010 when the company first started using MEMS microphones in the iPhone 4. Other important buyers of MEMS microphones include No. 2 Samsung and Chinese OEMs like Huawei, which also is in third place overall.
Amazon has become a heavy consumer of MEMS microphones, making up 45% of the total revenue of the MEMS microphones IoT segment. Outside of the handset space, Amazon is the biggest buyer, ahead of Hewlett-Packard. Amazon deploys up to eight MEMS microphones in its Echo smart speakers, which can be controlled through Alexa, Amazon’s digital assistant, via natural speech and voice command.