This Week In Bots: The Emotional, Eyebrow-Raising, Singing Edition

Robots are getting more expressive, more than being mere lumps of plastic designed to do jobs for us in handy ways. They can now almost sing…



The Moody Robo Butler

Mirai Sanzo is a simple little robot-esque device from Sanyo that’s closer to being a real product than many robot’s we’ve seen. He’s essentially a communication interface, powered by Android, that acts as a central point for a home automation system. He’s just 22cm across, has a touchscreen and touch sensors dotted around the edge, and does basic voice recognition so you can command your curtains closed and whatnot. 

What sets Mirai apart from other similar systems is his reactions: The static bot has lights inside so it can express up to seven “emotions.” This will let it warn you if your home’s consuming more energy than it’s making from solar cells or wind power, if your EV parked in the garage is running low on juice, or that you’ve rolled back in to your home late, reeking of beer and everyone else is asleep… Well, we’re presuming some of this because the little chap’s still pretty mysterious–despite the fact he’s going on sale in Japan soon.

Emotional Android

If you want to meet a much more emotional and human-like android, albeit one that’s far from coming to market, check out Emys from the Wroclaw University of Technology in Poland. He can read and give emotional reactions to people nearby. As a result, Emys can twist his odd-shaped face to replicate how humans look when feeling different emotions. Okay, so it’s an approximation…and while he’s far from the plastic features of the special effects Nexus 6 in Will Smith’s I, Robot, he’s still impressive. How impressive? Impressive enough that he gets his own amateur dramatic demo video that far surpasses the dry technical stuff that could’ve supported the project:


The Synthetic Singing Robo-Voice

Robots may increasingly look human, but no matter how realistic their synthesized voices get, they’re never totally convincing when they emanate from a loudspeaker. Which is something research by Kagawa University’s Hideyuki Sawada could, ultimately, change. What Sawada’s been working on is potentially a synthetic voicebox for robots, that uses very similar physics principles as our own vocal chords and mouths–ultimately producing what may be a more believable artificial voice. And now Sawada’s taught his quivering pink mass (see pic above!) to sing.

For a product demonstration that may have you clutching at your stomach, click below (unless, you know, you can’t handle it):


Despite the simple-looking nature of the tech, it’s actually very clever and monitors the sounds it’s making with a microphone so it can optimize them in real time (just as you do). If nothing else, the robot will remind you how fantastically complex your own body is, and that in the future we may find that we “trust” a robot that sounds human more than one that doesn’t–which could even have life-saving implications.

Chat about this news with Kit Eaton on Twitter and Fast Company too.

About the author

I'm covering the science/tech/generally-exciting-and-innovative beat for Fast Company. Follow me on Twitter, or Google+ and you'll hear tons of interesting stuff, I promise.