Google has long been considered a leader in the field of machine learning, and now it’s hoping to share some of its expertise.

The Web giant has just open-sourced a new machine learning system called TensorFlow that can do everything from finding matching images of your pet kitten to distinguishing your speech in a busy street.

The term “machine learning” sounds way more foreboding than it actually is. The concept works the same as how recognition works in humans – when you see someone you’ve seen in the past, your brain automatically recognizes them, even if it can’t always place their name. Google’s image search algorithm works in a similar fashion, and it can recognize photos of your face because they look similar. In this way, Google’s machines are effectively ‘learning’.

Now, Google’s machines are about to get a whole lot better. TensorFlow is a new system that’s not even in use yet, though the company plans to put it in place as soon as possible in every system that employs machine learning. This includes services like Smart Reply in the Inbox app, the search for similar images function in Google Images, and speech recognition in Google’s smartphone app.

TensorFlow can be thought of as a kind of artificial brain in which complex data structures, or ‘tensors’, flow through. Google says this process is a central aspect of deep learning, and can be used to enhance many of its products. The new system is said to be five times faster than Google’s current machine learning system, and can run on a single smartphone or literally thousands of computers.

In a blog post announcing TensorFlow’s release, Google’s CEO Sundar Pichai said the company is open-sourcing the technology in the hope it will accelerate research on machine learning and benefit the entire community. It’s also hoped the community can help to make the technology work better.

“We’ve seen firsthand what TensorFlow can do, and we think it could make an even bigger impact outside Google,” Pichai wrote. “We hope this will let the machine learning community — everyone from academic researchers, to engineers, to hobbyists — exchange ideas much more quickly, through working code rather than just research papers.”

Google has created the following video that shows exactly what TensorFlow can do: