John J. Smith, winner of last year's Readers Poll
begins a new series with this issue looking at the wider aspects of
computing.
SPEECH ... THE FINAL FRONTIER?
User-friendly is a term being thrown around at the
moment to describe how easy computers are to use but that is not how I
interpret this phrase. Imagine crawling out of your bed in the morning
and hearing a warm affectionate voice say "Good morning. Did you
sleep well? The kettle has boiled and the tea is ready. I await your
instructions". Later when you sit down at the keyboard instead of
the word READY a warm male or female voice (your choice) says
"What are we going to do today? Before we start I must tell you
that the office called to say that machine number 27 is on the blink
again. I have informed the repair man and he should be there just
after lunch". The computer would then stay quiet whilst you got
on with your writing or programming unless you asked it a question and
wanted a verbal reply. Now that's what I call User-friendly!
You may think that the above example is far fetched
but speech synthesis has come a long way over the past 5 years and
although they still have some way to go there are now devices
available which are quite intelligible provided they are programmed
properly.
In 1978 I did some manual searching of library
material for information on speech synthesis and speech recognition
(although I have never been involved in the hardware for this purpose)
and PAGE 6 readers may be interested to know that the earliest
reference I could find was dated 1952 in the Journal of Acoustic
Society of America Vol.24. However I understand that North
Staffordshire Polytechnic did a study in 1973 and that references have
been found on speech synthesis dating back to 1779(!) but these must
have been mechanical and are therefore hardly relevant today.
Nevertheless it is interesting that even with today's technology it is
still not perfected. It seemed at the time I was looking into this,
and is probably still true today, that more material was available on
speech synthesis - that is to get a computer to speak - than on speech
recognition - getting a computer to understand.
One interesting item was the work being carried out
by Sperry Gyroscope Inc. in the U.S.A .They had apparently built a
device called SCEPTRON which stands for Spectral Comparative Pattern
Recogniser. Other companies were working on this project as well for
the U.S. Navy and surprisingly the device was used to try and
understand the 'speech' of Dolphins. An article can be found on this
in Radio Electronics magazine (U.S.A) April 1964.
RCA and others were working on a mechanical interface to change speech
into a typewritten message in 1962 and Japan were working on a
transistorised multilingual speech to typewriter message system also
in 1962. In 1963 the reverse - written messages to speech - was being
worked on.
People have studied and experimented for many years
to make machines that talk but it is only with the advent of the
computer and in particular the large scale integrated circuit that
speech synthesis has been available to the home user in a reasonable
size package and at reasonable cost.
Texas Instruments took the rest of the industry by
surprise when it brought out a toy called Speak & Spell and
although it appeared to be an expensive toy at the time (was it really
7 years ago?) it was a miracle of modern technology. Several people
who were already into computing as a hobby bought one only to discover
that Texas would not provide any information on the chips nor would
they sell them on their own. Personally I think that was a big mistake
as, given the ingenuity of some hobbyists, I feel that this subject
could have advanced even quicker.
FURTHER AHEAD
For many years I have been saying that one day it
would be possible to dial up anything that can be read, watched or
listened to. If you can now download a piece of software or a news
item (via Teletext) it will not be long before you can get a print of
that book you wanted or that pop tune or the latest film, all by
dialling a phone number and downloading it. Now I am not talking about
recording the data on disk or tape or any other mechanical device
which must be subject to wear and tear but direct into computer
memory. Before long, with the cost per bit of memory coming down as
they pack more and more into each chip, I think we shall see
incredible amounts of non-volatile memory being used in our computers
and peripherals. We may gasp at hearing that the latest micro has half
a megabyte of RAM but as we get used to more RAM remember that
Operating Systems get more sophisticated with 68k Operating Systems
now becoming commonplace. If we want instant recall of what we want
when we want it then tens of thousands of megabytes are going to be
needed and probably several dedicated 32-bit chips processing in
parallel. Looking even further ahead a data highway could be built
into new houses controlling all our needs from straight information
retrieval to adjusting the central heating to automatically adjusting
the light coming through the window with some kind of
electro-luminescent material that would compensate for lack of natural
light and maintain a constant room light - with a manual override of
course! Main services such as gas, water, electricity, phone and data
charge would be read by remote computer accessing each of our systems
and billing us accordingly from a credit account.
GOVERNMENT CONTROL?
As time goes by I believe our computers will not
only get smarter but because of the sheer numbers involved and the
possibility of unlimited access to information they will become
difficult to control. Sooner or later someone will decide that each
computer will have its own built in identification code put there by
the manufacturer so that, when it goes on line, the Department of DATA
will have a source of income. Lets face it if people gave up smoking
and bought electric vehicles, revenue would have to come from
somewhere! I hasten to add that I am not recommending this as a course
of action!
I think that computing today is as exciting now as
in the beginning but there is still a lot more to come.
top
