As artificial intelligence and machine learning become more ubiquitous, the demand for power-efficiency increases.
Eliminating the power costs of digitisation, digital processing, and the transmission of irrelevant data becomes a whole new challenge. Even more so when consumers are growing increasingly fond of environmentally-friendly and sustainable brands.
Which includes Gen Z.
To combat this, a new chip is being introduced by Aspinity that reduces the amount of power needed for always-on ML by 95%.
It will have a major impact on the future of AI, ML, IoT and its implementation. Especially around new innovations for voice-enabled systems, home and commercial applications, and even healthcare.
Say Hello To The AML100 Analog Machine Learning Chip
Today’s always-on devices are pretty inefficient. They collect too much data, but don’t have enough power to process all of it.
Large volumes of irrelevant analog data ends up going through digitisation. And mostly at the expense of massive amounts of system power processing. And while reducing the quantity and movement of data through a system is one of the best ways to reduce power consumption, our current devices are pretty incapable of such a task.
So Aspinity has developed a new analog machine learning chip, known as the AML100.
The AML100 delivers substantial system-level power-savings by moving the ML workload to ultra-low-power analog circuitry. This allows the system to stay powered on for longer periods of time, enabling longer battery life and more efficient overall operation.
It essentially allows the chip to determine data relevancy with a high degree of accuracy and at near-zero power. Which results in a more efficient overall operation.
Founder and CEO of Aspinity, Tom Doyle said:
“We’ve long realised that reducing the power of each individual chip within an always-on system provides only incremental improvements to battery life. That’s not good enough for manufacturers who need revolutionary power improvements.
The AML100 reduces always-on system power to under 100µA, and that unlocks the potential of thousands of new kinds of applications running on battery.”
key features of the chip include:
- Consumes less than 20µA when always-sensing
- Intelligently reduces quantity of data by up to 100x while the data are still in analog
- Features field-programmable functionality to address a wide range of always-on applications
- Leverages patented analog compression technology for pre-roll collection to maintain accuracy of wake word engine in voice-enabled devices
- Supports 4 analog sensors in any combination (microphones, accelerometers, etc.)
- Available in 7mm x 7mm 48-pin QFN package
Production for the chip is set for Q4 2022, where customers can evaluate the AML100’s functionality.
Become An ML-Driven Business Today
The development of this chip is a major step forward in reducing the cost and power consumption of digital devices.
This will have a huge impact on the future implementation of AI, ML, IoT and more.
At Teraflow.ai, we help our clients stay ahead of the curve in this rapidly changing tech-driven landscape.
Contact us today to learn how our Machine Learning Experts can help you take advantage of these latest advancements.
