Dogs are not known to be the most advanced communicators, so figuring out what they want based on a few noises and pleading looks can be tough. This problem is what inspired a team of developers to come up with PUPPI — a small device that utilizes tinyML to interpret your canine companion’s mood through vocal signals. Their project employs an Arduino Nano 33 BLE Sense and its onboard microphone to both capture the data and run inferencing with the model they trained using Edge Impulse. After collecting ample amounts of data for barks, growls, whines, and other noises, their model achieved an accuracy of around 92%. 

Once deployed to the physical device, the board continuously takes in new sound data and comes up with a prediction for what kind of noise it is. This data is then sent over Bluetooth® Low Energy to an app that displays what the board is hearing, along with lighting up the onboard LED as a secondary indicator. 

The PUPPI is a cool showcase of the power contained within edge ML devices, and it will be exciting to see increased granularity in the classifications as more data is added. You can read more about this project here on Hackster.io.