Google launched a fun little app called Teachable Machine back in 2017 that allowed users to play around with the idea of machine learning. It allowed users to train an ML model, and see it in action straight away, without needing any code.
This week, Google is launching Teachable Machine 2.0, and it’s much more powerful than before. And best of all, you can actually use Teachable Machine 2.0 to add AI to an actual product.
Teachable Machine 2.0 introduces the ability to export your machine learning models to websites, apps, and other places. This means you can now make and train an ML model using Teachable Machine 2.0, and then use that on a website, an app, or wherever you like to do cool things like object detection and image classification.
Teachable Machine 2.0 also adds the ability to recognize sounds and poses, so you can now make ML models that can detect the difference in sounds and poses. The upgraded tool does look way more powerful than before, while still being really easy to use. It is pretty cool to see Google making ML really accessible to users who may not otherwise have the resources to build their own ML models.