Catherine Lyon, a computer scientist at the University of Hertfordshire, with her team of fellow researchers, has created a three-foot tall robot that helps adults learn how kids actually recognize and convert babbles into speech. Dubbed as DeeChee, the iCub robot can break up the stream of sounds into actual babble phrases and stores them in its memory. When a human volunteer tried to teach syllables to the robot, it could recognize and pronounce the most frequently used colors like green and red after repeated attempts.
As Catherine goes on to conclude, the breakthrough endeavor revealed that kids tend to avoid learning simple words like ‘at’, ‘with’ and ‘of’ since these are spoken in different ways. However, complex (concrete rather) words like ‘house’ or ‘blue’ involve uniformity of speech and are recognized by infants easily. Well, how about adding references, i.e. connotations of each word or syllable, to the conclusion? Are not the hidden nuances of words responsible for helping kids understand and attempt uttering those sounds?
Anyway, Catherine says,
When we asked people to talk to the robot as a small child, it seemed to come quite naturally to them. When they talk to a bit of disembodied software, you don’t get the same response.
Paul Goldstein, a psychologist at Cornell University says,
If we really think we understand how infants learn, then we should be able to build a robot that can do it.
The humanoid resembles human (a boy rather). Maybe, it makes easier for researchers to maintain a sense of connection. Aside from mimicking human speech, the robot can also recognize words of appreciation and encouragement and articulate the same in a jiffy. Moreover, the bot could recognize feelings like involve attention, affection, and gratitude, and communicated them through gestures, which involve smile, blink, etc.
Via: Physorg
