Ever wondered what will it be like for a robot placed in a room full of people that are talking to each other in in the middle of all that someone is asking it to carry out a task? Imagine a supermarket where a supervisor is shouting across the aisle asking a robot that cleans floors to mop up a particular area! Will the robot be able to follow such instructions?
With the current technology we have in place, chances are that they might not be able to. That’s exactly what scientists in the UK have achieved. They have developed a new microphone system that could help future robots pay better attention to what they hear in crowded places.
The technology, the scientists say, will make it possible for robots to zoom in on conversations in noisy rooms and listen for instructions.
One of the scientists who is leading the project is Dr Patrick Naylor from Imperial College London. Naylor explains that the ability to to pick out particular voice or conversation in a crowded place is a real challenge for everyday devices like phones and hearing aids.
Humans have the extraordinary ability to ignore all the unwanted sounds and tune in to particular sounds, picking voices out of a noisy environment. However, artificial intelligence isn’t smart enough yet and that’s the reason currently available robots, phones and other such devices that use speech recognition aren’t able to hone in on a particular sound in a crowded place.
Naylor revealed that the newly developed system, which is being demonstrated at the Royal Society’s Summer Science Exhibition in London, uses 32 microphones placed around a sphere. The system measures the tiny differences in the time it takes sound to reach each of the microphones and the information analysed by a computer. Using this information, the system can tune into sounds coming from one particular spot.
The lead scientist explains that our ability to listen selectively is what helps us communicate in this noisy world and for robots to be able to communicate effectively with us they need to have the ability to listen to different parts of the soundscape going on around it and pick out important conversations
“It’s a big challenge to be able to take away all the extra noises completely, but this is a first step towards that”, Naylor adds.
