Listening to audio story using human level artificial
intelligence
Note: To make this website free to the public
please click on an ad to support my sponsors or you
can make a tax-deductable donation using Paypal
(click on the donation icon on the left).
The robot's conscious is
the voice in the head that tells the robot
information about the environment. In terms of
listening to an audio story, the robot's conscious
will fabricate a movie based on the audio words. This
fabricated movie is a primitive sequence of images
and flow charts to represent the meaning to words
listened to.
This video is silent
at the beginning because I wanted viewers to look at
the thoughts of the robot while listening to the
audio story. I slowed things down so that people can
see the actual images and sound that is formed by the
audio text. The second part of the video will contain
audio sound to show viewers how the robot thinks at
normal speed.
When a human robot is
built, this is how it will think when it listens to
audio words. The robot was taught by teachers in
elementary school to understand meaning to words and
sentences. When words are spoken, visual images (or 5
sense data) are activated and these activated
thoughts are known as the meaning to the words
spoken. words=movie and movie=words. For example, if
the robot listens to the words: the cat jumped over
the box, the movie sequence will be a cat jumping
over a box. The movie activates the words and vice
versa.
The learning of
meaning to words come from a simple lesson. Teachers
take a word and they give a picture to represent that
word. The students will associate the word with the
picture. In the future, when the student reads a
word, a picture pops up in his head. This picture is
the meaning to the word. When we deal with sentences,
movie sequences activate in the mind and these movie
sequences are constructed based on complex patterns.
These complex patterns are formed in the robot's
brain based on comparing similar examples.
My robot doesn't use:
language parsers, grammar rules, semantic networks,
relational graphs, bayesian's network, genetic
programming, predicate calculus, rule-based systems,
and common sense systems (such as CYC). My robot
doesn't use machine learning, whereby programmers are
required to input knowledge into the system. My robot
learns all information from teachers in school.
This is important
because when the robot answers questions from the
audio story, he will be answering questions based on
the fabricated movie and not the audio words. Some of
the data in the fabricated movie contains logic and
isn't generated by the meaning to a word/s. One more
thing worth noting is that understanding natural
language is a small part of human level artificial
intelligence.