Abstract:The recent ubiquitous adoption of remote conferencing has been accompanied by omnipresent frustration with distorted or otherwise unclear voice communication. Audio enhancement can compensate for low-quality input signals from, for example, small true wireless earbuds, by applying noise suppression techniques. Such processing relies on voice activity detection (VAD) with low latency and the added capability of discriminating the wearer's voice from others - a task of significant computational complexity. The tight energy budget of devices as small as modern earphones, however, requires any system attempting to tackle this problem to do so with minimal power and processing overhead, while not relying on speaker-specific voice samples and training due to usability concerns. This paper presents the design and implementation of a custom research platform for low-power wireless earbuds based on novel, commercial, MEMS bone-conduction microphones. Such microphones can record the wearer's speech with much greater isolation, enabling personalized voice activity detection and further audio enhancement applications. Furthermore, the paper accurately evaluates a proposed low-power personalized speech detection algorithm based on bone conduction data and a recurrent neural network running on the implemented research platform. This algorithm is compared to an approach based on traditional microphone input. The performance of the bone conduction system, achieving detection of speech within 12.8ms at an accuracy of 95\% is evaluated. Different SoC choices are contrasted, with the final implementation based on the cutting-edge Ambiq Apollo 4 Blue SoC achieving 2.64mW average power consumption at 14uJ per inference, reaching 43h of battery life on a miniature 32mAh li-ion cell and without duty cycling.

A Novel All Silicon Bone Conduction Microphone with Broad Bandwidth (100Hz~10khz)

Design and Implementation of a Novel Skull Vibration Sensing Module for Bone Conduction Microphone

Design and Implementation of Dual Pressure Variation Chambers for Bone Conduction Microphone

BC-VAD: A Robust Bone Conduction Voice Activity Detection

Enabling Real-Time On-Chip Audio Super Resolution for Bone-Conduction Microphones

A Soft Resistive Acoustic Sensor Based on Suspended Standing Nanowire Membranes with Point Crack Design

In-Ear-Voice: Towards Milli-Watt Audio Enhancement With Bone-Conduction Microphones for In-Ear Sensing Platforms

DESIGN AND IMPLEMENTATION OF A CMOS-MEMS MICROPHONE WITHOUT THE BACK-PLATE

Prototype Optical Bionic Microphone with a Dual-Channel Mach–Zehnder Interferometric Transducer

Design and implementation of a novel double-poly microphone with rigid acoustic receiver, flexible spring supporter, and harm backplate

Bandwidth and SNR Enhancement of MEMS Microphones Using Two Poly-Si Micromachining Processes

A Real-Time Dual-Microphone Speech Enhancement Algorithm Assisted by Bone Conduction Sensor

Design and implementation of a novel CMOS MEMS condenser microphone with corrugated diaphragm

A Novel Piezoresitive Microphone MEMS Sensor For Aerospace Applications

An Electronic Stethoscope Based on MEMS Acoustic Vibration Microphone

Implementation of the CMOS MEMS Condenser Microphone with Corrugated Metal Diaphragm and Silicon Back-Plate

A miniaturized statically balanced compliant mechanism for on-chip ultralow wide-bandwidth vibrational energy harvesting

Silicon Condenser Microphone with Round Membranes

Frequency‐Selective, Multi‐Channel, Self‐Powered Artificial Basilar Membrane Sensor with a Spiral Shape and 24 Critical Bands Inspired by the Human Cochlea

A New Measurement Microphone Based on Mems Technology

A Wearable Bone-Conducted Speech Enhancement System For Strong Background Noises