If you’ve ever wanted a robot to respond to your emails, you’re in luck. Today, Google is announcing a new feature to its ambitious Inbox app that will enlist the company’s artificial neural network to craft automatic replies to incoming Gmail messages. You’ll never need to type again!
Just kidding, you will. The so-called Smart Reply feature is still in a pretty nascent state with responses limited to three to six words that you can then edit and send. They’ll probably be basic things like “Ha. Very funny.”
On top of that limitation, the neural network doesn’t always know exactly what you want to say, so it will offer up three options. It builds these responses by using two separate neural networks—one to read your emails and understand what people are saying to you and another to generate the automated responses. (It remains unclear if you’ll be able to respond with badass acid trip art generated by the neural network, though that seems highly unlikely.)
This is nevertheless a pretty neat advance in the otherwise antiquated realm of electronic messages. Wired’s Cade Metz explains how Google’s artificial neural network is learning how to talk like a human:
The system uses what’s called a “long short-term-memory,” or LSTM, neural network. Essentially, this is a neural net that exhibits something akin to human memory. It can “remember” the beginning of an email as it’s parsing the end—and that helps it, on some level, understand this natural language. In a research paper published earlier this year, a team of Google researchers showed how this technology could be used to build a “chatbot” that can carry on a decent conversation (in certain situations).
The research seems like it’s full of potential, but it’ll take some time before you can program a chatbot to talk to your friends. That said, Google’s already using its neural network to filter out spam, and the Smart Reply feature will be up and running soon. The company will make an announcement later today, and we’ll update with details.
[Wired]
Images via Google