Seeed Wio Terminal Supported by Edge Impulse (and It Can Sniff Alcohol!)

Running machine learning (ML) models on microcontrollers is one of the most exciting developments of the past years, allowing small battery-powered devices to detect complex motions, recognize sounds, or find anomalies in sensor data. To allow any developer to build these tiny ML models we’ve partnered with Seeed Studio to bring support for their Wio Terminal to Edge Impulse.

Card image cap

The Wio Terminal is a very versatile embedded system, with a powerful Cortex-M4F MCU, an LCD display, motion sensors, WiFi and BLE radios, and a microphone. And thanks to the enclosure it can easily be deployed as a stand-alone product. In addition, it has two easily accessible Grove connectors, allowing you to connect a very wide variety of sensors (over 300 available!) directly to the system.

And, now you can use this system to build your machine learning models! Using Edge Impulse you can now quickly collect real-world sensor data, train ML models on this data in the cloud, and then deploy the model back to your Wio Terminal. From there you can integrate the model into your applications with a single function call. Your sensors are then a whole lot smarter, being able to make sense of complex events in the real world.

Seeed Studio has created end-to-end tutorials on building models based on the accelerometer (to detect gestures), and for the Grove Multichannel Gas sensor (to distinguish different types of alcohol), but it’s very easy to integrate other sensors with a few lines of code, leveraging the Grove connectors at the bottom of the device, and the Edge Impulse Data forwarder.

Card image cap
Card image cap

Classifying alcohol in Edge Impulse using the Wio Terminal and a Grove multichannel gas sensor

Gathering data, and creating a machine learning model is one thing, but you can now also run your trained model in realtime on the Wio Terminal. This makes the model run without an internet connection, minimizes latency, and runs with minimum power consumption. Through the Edge Impulse studio you can either export your model as an Arduino library - to integrate it in your sketches - or as C++ source code - to integrate with your own firmware project.

Inspired? Head over to the Seeed Wio Terminal getting started guide and start building your very first embedded machine learning model. We can’t wait to see what you’ll build!

Jan Jongboom is the CTO and co-founder of Edge Impulse. He probably owns more Grove sensors than you.


Are you interested in bringing machine learning intelligence to your devices? We're happy to help.

Subscribe to our newsletter