Filmed at https://2019.dotjs.io on December 5-6, 2019 in Paris. More talks on https://dotconferences.com/talks Even though we don’t really think about it, sound is a very rich source of information. Leveraging the properties of sound with machine learning, we can gain insights about an environment or an activity performed. Acoustic activity recognition has the potential to create new interactions and better smart systems, and in this talk, Charlie explores how to experiment with this technology in JavaScript, using the Web audio API and Tensorflow.js.