Page 45 - EE Times Europe Magazine | June2020

EE|Times EUROPE — Sensors Insert 45

MICROELECTROMECHANICAL SYSTEMS
AAC Banks on Edinburgh’s Talents, Opens

MEMS Microphone Design Center

By Anne-Françoise Pelé

icrophones are ubiquitous in all aspects of the entertain-
ment industry, and their penetration is largely due to the
steady demand for smartphones, wearables, hearing aids,
MAR/VR headsets, and other consumer electronics devices.
Seizing the opportunity, AAC Technologies Holdings (Shenzhen, China)
has opened a microelectromechanical systems (MEMS) microphone
design center in Edinburgh, Scotland.
The Edinburgh center completes AAC’s global network of R&D
activities with a primary focus
on MEMS microphones. In the
short term, “our objective is to
develop a family of micro-
phones that can meet the AAC’s MEMS microphone (Image: AAC Technologies Holdings)
demanding specifications of
smartphones,” said Colin Jen- building microphone solutions, including the processing inside the
kins, R&D director and on-site microphone. “We will move in that direction soon,” said chief strategy
lead. “The initial microphone officer David Plekenpol.
we are working on is a very
high-performance digital FAST-MOVING MARKET
microphone. The entire IP The microphone market continues its upward trend. For the 2018–2024
[intellectual property] of the period, Yole predicts a compound annual growth rate of 3%, from
microphone is owned by AAC, US$1.7 billion to US$2 billion.
and it will probably be ready The MEMS microphone market, which now represents about 70%
for production before the end of the total, will grow from US$1.2 billion in 2018 to US$1.6 billion
AAC’s Colin Jenkins of this year.” in 2024. Key driving markets include smartphones, smart speakers,
The center aims to and hearables, such as wireless earbuds. “In the last couple of years,
strategically position AAC at a time when voice-controlled artificial the smart-speaker and hearable markets have experienced explosive
intelligence applications are becoming increasingly central to the growth,” said Damianos. In volume terms, MEMS microphones in
smartphone and mobile user experience. It will focus on high-end smart speakers will grow at a CAGR of 13% to 1.2 billion units in 2024.
MEMS microphone devices with smaller physical dimensions, better In wireless earbuds, they will expand at a CAGR of 29% to 1.3 billion
signal-to-noise ratio (SNR), units in 2024.
lower power draw, and To reap the benefits of the ever-growing trend, MEMS companies are
improved ruggedness and trying to move up the value chain and become more vertically inte-
reliability. With such features, grated. “The ASIC, the transducer, and the packaging [constitute] the
MEMS microphones can be MEMS microphone, and if you have two out of three, then you rely on
used in devices that offer an outside partner to build, for example, the transducer,” said
improved keyword detection, Plekenpol. “Our intention is to have all three pieces basically coming
user recognition, multi- from AAC. That’s an ever-present roadmap of development for high-
language recognition, and SNR microphones.”
custom wake-up words. Recently, Infineon AG has shifted its business model from selling
“The added value of AI is for microphone dies to players such as AAC to selling complete packaged
the natural-language process- MEMS microphones, and it has expanded beyond MEMS microphone
ing,” said Dimitrios Damianos, manufacturing to become an integrated provider that handles man-
technology and market analyst ufacturing, packaging, testing, and sales of the devices. When asked
in the photonics and sensing about this, Plekenpol said he understood Infineon’s logic, because
division at Yole Développment that’s the way to go “if you want to develop a complete solution — not
AAC’s David Plekenpol (Lyon, France). “The voice is a only [building the] ASIC, transducer, and packaging but actually build-
more natural way to interact ing a system-level solution.”
with the machine. You don’t have to use a keyboard. You don’t have to He added, “If you have the resources and the talent to build all those
use your hands. You just use your voice.” But understanding a speaker’s elements on your own, then you have a more integrated product and, I
intent — not just detecting language but grasping context — requires think, a better financial performance.”
a lot of processing. “AI is adding the value of decoding and helping our
communication with our devices,” Damianos said. EUROPE’S WEALTH OF TALENT
AAC will work to optimize the signal pathway and to achieve power AAC intends to leverage R&D resources and pursue its expansion in
savings by limiting the need for a power-hungry DSP to process audio Europe over the next three to five years. “Not that we can’t find [talent]
signals. The company will focus on a more system-level approach to in Asia, and more specifically in China, but I suppose that for a global

www.eetimes.eu | JUNE 2020