Today the new BBC micro:bit was announced, the successor of the highly popular BBC micro:bit. While it retains the same form factor it now packs a microphone and a speaker, and a much more powerful MCU. And what better way to demonstrate these features than by adding some Machine Learning to the new micro:bit?!
In the video above you'll learn how to make your micro:bit respond to a keyword, by training a machine learning model that recognizes your voice. You can use this to make a voice-activated lock, turn LEDs on on command, or control a motor when you say a magic spell. All you need is your new micro:bit, a free Edge Impulse account, your phone (to record new data), and an hour of your time!
Here are some useful resources that are referenced in the video:
- Sign up for an Edge Impulse account
- A basic dataset of 'noise' and 'unknown' data, sampled at 11KHz
- Repository with source code to run the trained model on your new micro:bit
- Want to learn more about the underlying algorithms? See the tutorial on recognizing sounds from audio.
Built something interesting? Got questions? Let us know in the Edge Impulse forums.
We can't wait to see what you'll build! 🚀
*Jan Jongboom is the CTO and co-founder of Edge Impulse. For the launch of the first micro:bit he managed to get someone to juggle devices live on camera.*