MEMS microphones are still improving and continue to enhance the human-device communication – An interview with TDK InvenSense

When people gather in groups, they share stories, experiences, feelings; they communicate. The way to convey everything that is on people’s minds is by using their voice, the most natural form of communication. Now, voice is becoming more and more relevant in a digitalized world, helping people extend their communication with everyday devices. And for understanding the context of a user’s intent, audio capture devices (namely microphones), are becoming ever more relevant.

Yole Développement (Yole) is closely following this voice-based human-machine interaction trend, which is poised to boost the MEMS market, growing annually at a 5.4% CAGR2019-2025, surpassing the US$1.6 billion mark by 2025. The MEMS mic market is not expected to experience any strong negative effects from covid-19.

In that context, Dimitrios Damianos and Alexis Debray, both PhD. Technology & Market Analyst at Yole, have interviewed Kieran Harney, Director of Audio Products at InvenSense (TDK group company), a leading MEMS player offering MEMS microphone solutions, among other devices.

Dimitrios Damianos (DD): TDK InvenSense is a world leader in MEMS. Please provide a short overview of your microphone business as of today. What is your role/responsibilities, in particular?

Kieran Harney(KH): InvenSense, a TDK group company has a broad portfolio of sensor products, including MEMS microphones, six axis motion sensors, pressure sensors, ultrasonic proximity and fingerprint sensors. These products are sold across a wide range of markets including mobile phones, consumer and IoT, robotics, automotive and industrial. We offer a range of analog and digital output performance-based microphones with world leading SNR, AOP and power. Our microphones come in both top and bottom port configuration and in very small form factors. I am responsible for marketing and business development in our Microphone Business Unit.

DD: At Yole, we have seen TDK release many MEMS products recently, including the newest T3902 PDM microphone. Could you tell us a little bit more about this product (what is new, etc.)? How did you achieve an ultra-low power consumption?

KH: We put a lot of investment into new product development and our focus is on creating value for our end customer though performance and features. The T3902 utilizes an ASIC that was designed with low power in mind. We have a world class microphone ASIC and MEMS design team that worked together to optimize the MEMS performance and the ASIC design to deliver a low-power mode power consumption of 185µA, which is ideally suited for always-on applications where battery life is a premium. The product comes in a compact 3.5mm x 2.65mm footprint and is a great fit for mobile phones, smart speakers, true wireless sound headsets, TV remote controls and many other applications.

We (TDK InvenSense) put a lot of investment into new product development and our focus is on creating value for our end customer though performance and features.

Kieran Harney

Managing Director of Audio Products at InvenSense

DD: The MEMS microphone business is special, with its low margins, sometimes, and high quantities shipped annually. We are seeing a lot of differentiation: either on the material level (piezoelectric tech for example), the transducer/sensor level (for example dual membrane), and the ASIC level (extra intelligence on edge). In your case, what are the differentiating factors and what advantages do they bring?

KH: At TDK, we are very focused on innovation and differentiation. Our core MEMS technology achieves the highest SNR and the highest AOP performance available in the market today. This technology includes multi-membrane and multi-backplate sensor designs. Our ASIC designers have achieved the lowest power analog and digital output microphone while maintaining superior electroacoustic performance. Utilizing these technology cores, we have been able to create a wide range of market leading products.  Our 74dB SNR analog output MEMS microphone is the lowest noise MEMS microphone available on the market today. Our T5818 PDM MEMS microphone has a very wide dynamic range of 107dB with an acoustic overload point of 135dB SPL. The T5808 is the world’s first SoundWire compatible MEMS microphone product designed to be compatible with Qualcomm SoundWire enabled chipsets. We recently announced the ICS-40638 that supports 138dB AOP and is rated for 105°C.   The ICS-40800 is the world’s first directional MEMS microphone. Very high SNR, AOP and ultra-low power are the leading requirements for microphone customers. Using our microphone technology, customers can simultaneously achieve high quality audio capture, multi-mic beamforming, voice recognition, low-power always-on and active noise cancellation in very challenging environments. 

Courtesy of TDK InvenSense, a TDK group company, 2020

Alexis Debray (AD): We have seen for quite some time the rise of voice as a basis for human-machine interaction. We expect this trend to continue, as voice is a seamless way of interacting with everyday devices. Voice Personal Assistants are becoming standard in smartphones, smart speakers, wearables & hearables, which in turn raises the requirements for MEMS mics (SNR, AOP, beamforming, ANC, low-power, always-on, etc.) How do you see your MEMS microphones fitting into this, let’s say, VIoT (Voice Internet-of-Things)?

KH: IoT / VIoT covers a very diverse and growing market with many different demands on microphone performance, interface and form factor. The advent of MEMS microphones has enabled audio designers to create products with performance levels unthinkable using traditional Electret Condenser Microphone (ECM) technology.  Audio system designers can now leverage the part-to-part consistency, the performance density and the stability over time of MEMS microphones to create leading edge voice enabled technologies. TDK has the widest portfolio of MEMS microphones on the market today, tailored to meet these design challenges. 

The table below summarizes some of our market leading capabilities.  

Highest SNR 74dBA, ICS-40730
Ultra-High SPL 138dB SPL, ICS-40638
Lowest Power Analog 16µA, ICS-40310
Lowest Power Digital 185µA, T3902
Most Interfaces Analog – Single Ended & Differential Digital  – PDM, I2S, TDM and SoundWire
Ultra-Small Package 2.7mm x 1.6mm x 0.9mm, T4064
Directional Tunable response from cardioid to figure eight, ICS-40800

AD: Do you see other emerging markets and applications for MEMS microphones?

KH: True Wireless Stereo (TWS) is emerging as a major growth driver for MEMS microphones.   Recent technologies are driving as many as 4 or 5 microphones per ear bud. This is a great example of a market feature that has been enabled by the performance density of MEMS microphones.  In automotive markets, we see rapid adoption of multiple microphones to support voice capture, voice command and active voice cancellation for road noise. 

AD: Is there any specific microphone product/technology roadmap that you could share? Or, at least, what are your next steps? Does it include other technologies, better ASIC design, edge computing, pre-processing, or algorithm development?

KH: MEMS microphones today are passive sensors with minimal configurability, where electroacoustic performance has been optimized for the application. However, these microphones are at the heart of audio capture for voice recognition. There is a lot of focus on trying to push processing to the edge and ultimately into the microphone itself. Our focus is in identifying what level of capability belongs in the microphone to support an optimized processing hierarchy. Pushing too much into the microphone creates high complexity and cost for implementation in a system. Too little does not offload anything from the application processor. Our roadmap is focused on finding the correct balance of technology inside the microphone itself. 

We believe that the market is looking for a better way to achieve very low power voice detection and voice wake up. Limited solutions today are either complex high-power duplications of application processor functions or are low-power with compromised acoustic performance and unreliable detection. Our development activity is focused on facilitating an optimum voice activation solution while driving industry leading power and acoustic performance. 

Interviewee

Kieran Harney is Managing Director of Audio Products at InvenSense, a TDK group company.

Based in Boston, Massachusetts, he is responsible for marketing and business development for MEMS microphones.  Before joining InvenSense, Kieran was Product Line Director for MEMS microphones at Analog Devices, Wilmington, Massachusetts. He created a successful microphone business at ADI and led the sale of this business to InvenSense in 2013. Prior to this, he was responsible for assembly and test operations and engineering development for MEMS accelerometers and gyroscopes at Analog Devices facility in Cambridge, Massachusetts. Kieran also held roles as assembly engineering manager at Analog Devices Philippines and as assembly packaging engineer at ADI in Ireland. He holds over 40 US patents in the area of MEMS sensor packaging and MEMS Microphones. 

Kieran received a B.Eng. in Manufacturing Engineering and an MBA from the University of Limerick, Ireland.  

Interviewers

Dimitrios Damianos - Yole Développement

Dimitrios Damianos, PhD joined Yole Développement (Yole) as a Technology and Market Analyst and is working within the Photonics, Sensing & Display division.

Dimitrios is daily working with his team to deliver valuable technology & market reports regarding the imaging industry including photonics & sensors.

After his research on theoretical and experimental quantum optics and laser light generation, Dimitrios pursued a Ph.D. in optical and electrical characterization of dielectric materials on silicon with applications in photovoltaics and image sensors, as well as SOI for microelectronics at Grenoble’s university (France).

In addition, Dimitrios holds a MSc degree in Photonics from the University of Patras (Greece). He has also authored and co-authored several scientific papers in international peer-reviewed journals.

Alexis Debray - Yole Développement

Alexis Debray, PhD is a Technology & Market Analyst, Optoelectronics at Yole Développement (Yole). As a member of the Photonics, Sensing & Display division, Alexis is today engaged in the development of technology & market reports as well as the production of custom consulting projects dedicated to the imaging industry.

After spending 2 years at the University of Tokyo to develop expertise focused on MEMS technologies, Alexis served as a research engineer at Canon Inc. Over 15 years he contributed to numerous projects of development, focused on MEMS devices, lingual prehension, and terahertz imaging devices.

Alexis is the author of various scientific publications and patents. He graduated from ENSICAEN and holds a PhD in applied acoustics.

Related reports

Microphone, Microspeaker and Audio Solution Market and Technology Trends 2019

Microphones, Microspeakers and Audio Solutions Market and Technology Trends 2019

The voice-based world is shaking up the audio industry, making it worth $20.8B in 2024.

AirPods SiP Audio Module back side chip - System Plus Consulting

Advanced System-in-Package Technology in Apple’s AirPods Pro

Analysis of Apple’s first SiP found in the latest AirPods, featuring a fully integrated SiP for audio codec and Bluetooth connectivity.

Source: http://www.yole.fr

Related presentations

Liked this post?

Share it on your social networks

Photonics Forums Yole CIOE 2020 1140x370