“Active Noise Cancellation (ANC) is an ideal feature for wireless earbuds, which successfully reduces background noise and wind noise, but also reduces sound reception.Due to this issue, ANC is often turned off during voice calls[1]. To overcome this limitation, a new type of MEMS device, called a “Voice Pickup Sensor (VPU)”, captures sound directly through bone and skull vibrations[2]or “Voice Pickup Accelerometer”[1]referred to as “accelerometer”[3]consisting of a MEMS accelerometer with MEMS microphone frequency characteristics.
“
Author: Dr. Arnaud Parent, Senior Application Engineer, Coventor (a company of Lam Group)
In our previously published article, “Understanding MEMS Microphone Technology and Design,” we pointed out that the demand for MEMS microphones is driven by customer-requested audio capabilities, such as stereo sound, voice recognition, sound directivity, noise cancellation, and more. In order to achieve these audio characteristics, multiple MEMS microphones are required to support the function in the final product. For example, state-of-the-art smartphones may contain as many as 6 microphones. Due to the increasing demand for remote work, the demand for MEMS microphones has also accelerated to a certain extent. Telecommuting and online meetings are becoming the new normal, and these meetings would not be possible without a good voice input system (i.e. microphone).
Voice sensors are being used in True Wireless Stereo (TWS) earbuds. These earbuds connect wirelessly and connect with smartphones or other Bluetooth devices. Each stereo earbud contains a speaker (possibly a MEMS microspeaker in the future) and multiple MEMS microphones. In most advanced TWS systems, each earbud contains three MEMS microphones: two external microphones for beamforming and one internal microphone for active noise cancellation (ANC). Because there must be some distance between the two external microphones, a typical P-type TWS earbud will have a long bar (as shown in Figure 1).
Figure 1: Commercialized TWS earbuds (Image credit: Apple)
Active Noise Cancellation (ANC) is an ideal feature for wireless earbuds, which successfully reduces background noise and wind noise, but also reduces sound reception.Due to this issue, ANC is often turned off during voice calls[1]. To overcome this limitation, a new type of MEMS device, called a “Voice Pickup Sensor (VPU)”, captures sound directly through bone and skull vibrations[2]or “Voice Pickup Accelerometer”[1]referred to as “accelerometer”[3], consisting of a MEMS accelerometer with MEMS microphone frequency characteristics. Basically, the device is a hybrid of a microphone and an accelerometer. A simplified VPU sensor consists of a modified MEMS microphone (with a mass attached to the microphone’s diaphragm) and an airtight sound port (Figure 2).
Figure 2: Schematic of a MEMS microphone (left) and a voice pickup (VPU) bone sensor (right). Note the added mass and missing sound port on the VPU.
In addition to improved speech detection and background noise immunity, the VPU has some other advantages in active noise reduction applications:
● Support for tap detection: This provides an interactive way to control the device, such as one tap to increase the speaker volume and two taps to decrease the volume;
● Voice Interruption: VPU picks up voice activity even when music is playing from the internal speakers, which means that conversation can still be heard even when listening to loud music;
● Since no sound port is used or required, the VPU can be completely sealed to better protect it from external environmental contamination.
Figure 3: MEMS+ VPU model (1/4 of the device shown)
MEMS+® is a tool of the Coventor MP® MEMS design platform for designing MEMS microphones, accelerometers and VPU sensors. The MEMS+ VPU model is shown in Figure 3. The model accurately reproduces the nonlinear behavior of the device, which is critical in circuit and system-level design. At the same time, the model is fully parametric and supports multiphysics modeling. The latest version of the MEMS design software, CoventorMP 2.0, has also been released. This release allows charging sources to be used in MEMS designs, which is key to testing MEMS microphones and VPUs in real-world configurations. Every feature in CoventorMP can be used to design and develop highly advanced next-generation speech detection equipment.
The demand for MEMS microphones will continue to increase as new applications and new products with voice sensing come out. New MEMS-based voice sensors (such as VPUs) that support advanced features, and the need for new and better voice-sensing solutions will continue to drive the next generation of MEMS microphones and MEMS-based voice-sensing products.
References:
[1] Vesper white paper, Voice Accelerometers in TWS Earbuds, A Sensor and Algorithm Perspective
[2] Paul Clemens, Sonion Voice Pick up (VPU) Sensor, TDK Developers Conference 2018, Santa Clara Marriott, September 17-18, 2018
[3] ST-Microelectronic brochure, LIS25BA, Low-noise, high-bandwidth accelerometer with TDM interface
Original link: https://www.coventor.com/blog/make-voice-heard-latest-advances-mems-voic…
The Links: LTM150XH-L04 CM1000DU-34NF