SynSense advances ultra-low-power neuromorphic audio processing with Xylo™Audio 3 tapeout
Based on temporally sparse, quantized neural networks (spiking neural networks, or SNNs), Xylo™Audio 3 is an ultra-low-power always-on AI chip for auditory processing. The highly efficient SNN temporal processing units in Xylo™ reduce the energy consumption for complex temporal signal analysis tasks by at least 10 times.
Xylo™Audio 3 integrates a configurable audio front-end (AFE) which accurately converts audio inputs into spikes, with excellent compatibility for a range of analog and digital MEMS microphones.
The latest enhancements to the Xylo™Audio family feature more flexibility in network configuration coupled with lower-power operation. Xylo™Audio 3 enables edge AI for near-sensor auditory applications, for keyword recognition, environmental sound monitoring, industrial vibration monitoring and more, empowering a wide range of edge intelligence applications.
“We are providing Xylo™Audio to many commercial and research partners, who are developing applications ranging from in-home audio event monitoring to underwater acoustic signature detection. Xylo™ enables our partners to build sophisticated audio-driven intelligence into their battery-powered edge devices,” said Dylan Muir.
In addition to its innovative chip design, Xylo™Audio 3’s use of the TSMC 40nm process contributes to cost reduction and ensures the stability of tapeout and production. It allows SynSense to deliver cost-effective, high-performance neuromorphic audio processing technology that is applicable across various industries.
SynSense has already launched development kits based on Xylo™Audio, attracting considerable attention from the research and industry sectors. The company is actively supporting customers in application development. Muir added, “We expect engineering samples of Xylo™Audio to be available at the end of 2023, with commercial volume sales to follow.”
SynSense is a world-leading neuromorphic engineering company. SynSense provides custom-tailored, ultra-low-power silicon design solutions for industrial and consumer machine-learning inference applications. As a “full-stack” neuromorphic engineering company, SynSense delivers complete solutions, including custom IP, hardware, and software configurations to meet specific application needs. SynSense was founded in March 2017 in Zürich Switzerland, based on ground-breaking advances in neuromorphic computing hardware developed at the Institute of Neuroinformatics of the University of Zurich and the ETH Zurich.
Dylan Muir, email@example.com
Nancy Huang, firstname.lastname@example.org, 028-62073881