City/Region: Pittsburgh


Funding Series: -
Employees: 11-50
Engineers: Unknown
Year Founded: 2015
Connect With Us

About the Company

Aspinity provides ultra-low power, trainable and programmable analog machine learning (AnalogMLTM) solutions that extend battery life in a wide range of portable, always-on sensing applications ranging from Iot and smart home to hearables and wearables, heart rate sensors, and industrial vibration monitors.

How is the team at Aspinity adapting to COVID-19?

In many ways, COVID has accelerated the need for portable voice enabled devices for hands free operation. While their team is mostly working remotely, they have not slowed down operations.

Featured Content

Aspinity RAMP voice activity detection with preroll







Reduce the Data, Save the Battery

Aspinity CEO Tom Doyle explains how intelligently minimizing the amount of data running through an always-listening system preserves battery life.

Product Description and Tech Stack

Aspinity Analog Machine Learning (AnalogMLTM) Chip

An AnalogML chip uses near-zero power to detect events such as voice, glass break, a vibration change, or a heart rate anomaly while sensor data are still in their natural analog domain, eliminating the wasteful digitization and downstream processing of irrelevant data. By focusing the system power usage on the important data and keeping the digital components of the system off unless an event is detected, an analogML chip delivers a 10x improvement in battery life in portable, always-on sensing devices. Learn more about how Aspinity saves battery life in always-on device here:

What is the technology behind AnalogML?

AnalogML is enabled by Aspinity’s innovative and patented RAMPTM (Reconfigurable Analog Modular Processing) technology platform, a scalable and programmable analog signal processing technology (ASP) capable of replicating sophisticated digital signal processing tasks in analog circuitry at much lower power. RAMP technology leverages the large-signal characteristics of a small number of transistors to develop an architecture of modular, parallel, and continuously operating analog blocks that are configurable for typical analog tasks such as sensor interfacing, signal processing, and data conversion in addition to more complex tasks such as feature extraction and inferencing, leading to a significant improvement in power efficiency for the entire system.

Technical Leadership Team


Work Policy and Location

Work Policy: Fully Remote

2000 Smallman St Ste 201 Pittsburgh, PA 15222

Photo Gallery