The past years have brought us three really interesting things: 1) TensorFlow Lite, a portable runtime for neural networks aimed at running machine learning models on mobile devices. 2) Emscripten, a toolchain that can compile C and C++ applications into WebAssembly, making these applications available from any browser. 3) Access to many sensors straight from your mobile web browser. All these features are cool on their own, but together they enable something truly magical: classifying what happens in the real world straight from the browser. Let's put that in practice!
This is a companion discussion topic for the original entry at https://www.edgeimpulse.com/blog/audio-based-shower-timer-with-a-phone-machine-learning-and-webassembly/