Sized at just 24mm x 28mm, the Syntiant TinyML board is a small, self-contained system that allows trained models to be easily downloaded via Edge Impulse through a micro-USB connection without the need for any specialized hardware. Equipped with Syntiant's ultra-low-power NDP101 Neural Decision Processor, the TinyML board can enable speech and sensor applications to run at under 140 and 100 microwatts, respectively, delivering 20x more throughput and 200x efficiency improvement compared to traditional MCU-based systems. The new board also is fully compatible with Arduino’s open-source platform.
"Syntiant’s TinyML board is another example of how we are advancing AI pervasiveness by moving machine learning from the cloud to the edge,” says Kurt Busch, CEO of Syntiant. “Our collaboration with Edge Impulse now enables anyone, from individual developers and hardware engineers to small companies to design, build and deploy highly accurate ML applications that respond to speech, sounds and motion with minimal power consumption. Whether it is for a wearable, industrial product or even to assist with people with disabilities, the possibilities are endless with our new TinyML board that provides a full solution for bringing the power of artificial intelligence to almost any device.”
Users can easily sample raw data, build, download and deploy trained embedded machine learning models by using the Edge Impulse studio to create low-power, high-performance speech, audio and sensor interfaces.
"In Edge Impulse's pursuit to democratize embedded machine learning for all developers, we are thrilled to partner with Syntiant, which has become a trailblazer in the ultra-low-power edge AI space,” says Zach Shelby, co-founder and CEO of Edge Impulse. “Our front-end support for the Syntiant TinyML Development Board provides customers with all the software tools necessary for building and deploying limitless ML applications that will unlock the next generation of always-on devices."
The TinyML platform’s onboard digital microphone and motion sensor enable easy configuration for any speech, event detection or 6 axis motion- and vibration-related application. Trained models can be easily downloaded on the TinyML board through the USB connection without the need for any specialized hardware. The board is shipping with a built-in “Alexa play music” model, ready for experimentation.
Complementing the Syntiant NDP101 Neural Decision Processor, developers will be able to leverage Syntiant's Core 1 neural network with support for up to 560k parameters and 64 output classifications, in combination with a ARM Cortex-M0+ 32bit host processor with 256KB of flash memory and 32KB host processor SRAM, plus 2MB on-board serial flash memory. The 5 digital I/Os are compatible with Arduino MKR series boards.
www.edgeimpulse.com
www.syntiant.com
