Wednesday September 19, 2018
Home Lead Story Google AI can...

Google AI can focus on individual speakers in a crowd

The visual signal not only improves the speech separation quality significantly in cases of mixed speech, but, importantly, it also associates the separated, clean speech tracks

0
//
38
Google india launches 'Tz' to help people pay their utility bills. Wikimedia Commons
Google AI to identify speakers from crowd. Wikimedia Commons
Republish
Reprint

Just as most smartphone cameras now allow users to focus on a single object among many, it may soon be possible to pick out individual voices in a crowd by suppressing all other sounds, thanks to a new Artificial Intelligence (AI) system developed by Google researchers.

This is an important development as computers as not as good as humans at focusing their attention on a particular person in a noisy environment. Known as the cocktail party effect, the capability to mentally “mute” all other voices and sounds comes natural to us humans.

Google has collaborated with getty images. Wikimedia Commons
Google AI will identify individual speakers now. Wikimedia Commons

However, automatic speech separation — separating an audio signal into its individual speech sources — remains a significant challenge for computers, Inbar Mosseri and Oran Lang, software engineers at Google Research, wrote in a blog post this week. In a new paper, the researchers presented a deep learning audio-visual model for isolating a single speech signal from a mixture of sounds such as other voices and background noise.

“In this work, we are able to computationally produce videos in which speech of specific people is enhanced while all other sounds are suppressed,” Mosseri and Lang said. The method works on ordinary videos with a single audio track, and all that is required from the user is to select the face of the person in the video they want to hear, or to have such a person be selected algorithmically based on context.

Also Read: Want To Know What Facebook, Google Know About You?

The researchers believe this capability can have a wide range of applications, from speech enhancement and recognition in videos, through video conferencing, to improved hearing aids, especially in situations where there are multiple people speaking. “A unique aspect of our technique is in combining both the auditory and visual signals of an input video to separate the speech,” the researchers said.

google
This will also help in speech enhancement . VOA

“Intuitively, movements of a person’s mouth, for example, should correlate with the sounds produced as that person is speaking, which in turn can help identify which parts of the audio correspond to that person,” they explained.

The visual signal not only improves the speech separation quality significantly in cases of mixed speech, but, importantly, it also associates the separated, clean speech tracks with the visible speakers in the video, the researchers said. IANS

Click here for reuse options!
Copyright 2018 NewsGram

Next Story

Why Robots That Are Made To Look Like Human Make Them Feel Uneasy

People who design machines to work with humans do keep the uncanny valley in mind as they think about the look of a robot.

0
Robots
Why Robots That Look Too Human Make Some People Uneasy Pixabay

An increasing number of robots are being created and designed to work side by side with humans, in a human environment. That means robots have to be structured like a person, because some of them have to walk and sit like a person. Some robots are even being designed to look human.

But seeing an android, a robot that looks human, can make some people uneasy. That growing unsettling feeling or phenomenon as robots begin to look more like human beings is called the “uncanny valley.”

Even researchers who work on robots are not immune to it.

“I know how they work. I know they’re just machines, but something about something that looks like a person but doesn’t quite move like a person is disturbing,” said Jonathan Gratch, director for virtual human research at the University of Southern California’s (USC) Institute for Creative Technologies.

Gratch, who is a research professor of computer science and psychology, studies human-computer interaction.

He said there are many thoughts behind why the uncanny valley exists. One explanation is that it’s biological. People are hardwired to recognize when something seems wrong.

“In my research, I study emotion and how we use emotional cues to read each other’s minds, and I think a lot of the issue for me is if you try to make something very realistic, then you start trying to read all this information into what it’s portraying, and it is not the right information. So, it just communicates something is off. Something is wrong with this interaction,” Gratch said.

Another theory is that a robot that looks too human threatens what it means to be human.

robots
A robot head is covered by Hanson Robotics’ skin, in Hong Kong. VOA

“Initially, humans were seen as the only intelligent entity. And now, we know more and more that animals can do many of the things that we do, build tools. We know machines are starting to become intelligent. We hold on to the fact that we’re emotional, but now these machines are starting to be emotional as well, which is perhaps a threat. So, where does that lead people?” Gratch explained.

A person’s religious beliefs and culture may also play into how an android is perceived, he suggested.

“In the Western tradition, coming from Christianity, humans are unique, perhaps uniquely possessing a soul. Whereas in Japanese Shinto culture, souls live everywhere, in rocks and machines,” Gratch said.

John Rebula is a postdoctoral fellow at USC and is working on making a humanoid robot walk like a person by being more coordinated and balanced. Applications include the ability to walk up a flight of stairs and sit in a chair made for a person. He said the robot’s face is not necessary and is clearly cosmetic.

Technology, robot
The mock killer robot was displayed in London in April 2013 during the launching of the Campaign to Stop Killer Robots, which calls for the ban of lethal robot weapons that would be able to select and attack targets without any human intervention. (VOA)

“We really do think of these as research machines that we’re ripping apart and putting back together, ripping apart and putting back together. And so, it’s very easy for us to leave off the cosmetic bits,” Rebula said.

His robot does have cartoon-like eyes, ears and a nose. It could be considered cute. However, if it looked more human, Rebula said he would not necessarily want to be in the lab with it all the time.

“We have lots of late nights in labs. You start yelling at the robot a little bit as it is — ‘Oh, why aren’t you working?’ I don’t necessarily, myself, need that extra layer of weird,” Rebula said.

Also Read: Video- India Scraps Law Criminalizing Homosexuality

People who design machines to work with humans do keep the uncanny valley in mind as they think about the look of a robot, and how widely it will be accepted by humans. (VOA)