Aspinity is a semiconductor company specializing in analog machine learning (ML) technology for edge devices. The company's core product is the analogML core, a neuromorphic computing architecture designed to significantly reduce power consumption in always-on sensing applications. Aspinity's technology addresses the challenge of battery drain in voice-enabled gadgets and other smart devices that are constantly listening for wake words or specific acoustic triggers.
The analogML core uses an "analyze-first" approach, evaluating data relevance in the analog domain before digitization. This method allows the core to process raw, unstructured sensor data without relying on power-hungry digitization and digital processors. By eliminating extraneous data at the beginning of the signal chain, Aspinity claims its technology can enhance battery life by 10 times or more in applications such as voice activity detection, acoustic event detection, and vibration monitoring.
In March 2022, Aspinity introduced the AML100, which reduces always-on system power to under 100μA. The company has since announced plans to move to a 22nm process for its AML200, aiming to deliver TOPS (Trillions of Operations Per Second) level performance within a battery-operated power footprint. Aspinity has also expanded its focus to include automotive security, launching a new suite of algorithms and a dashcam evaluation kit.
Key customers and partnerships
Aspinity has formed partnerships with major players in the semiconductor industry. In December 2019, the company demonstrated an ultra-low-power analog voice wake-up system using STMicroelectronics' microcontrollers. In May 2020, Aspinity announced a collaboration with Infineon Technologies to accelerate the development of battery-operated always-on sensing products for consumer and IoT applications, utilizing Infineon's XENSIV family of sensors.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.