Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Measureable differences between phomemes - where should a layman read?

10 views
Skip to first unread message

Richard Owlett

unread,
Sep 14, 2011, 5:07:08 PM9/14/11
to
I don't know how to phrase the question. So I'll give to examples.

1. Back in the early 60's a one semester intro to linguistics at
Cornell. I well remember one demonstration of allophones vs
phonemes. The vast majority were from Upstate New York or from New
England. Two fellow were from the same town in the Deeeep South. The
class unanimously agreed they were the people with a pronounced
accent ;)

The instructor asked them to trade off saying "pin" and "pen". He
had set this up before class so the class and he did not know in
what order/permutation the words would be used. He and the
non-speaker could always correctly identify what was being said. The
rest of could never say that different words were being said.


2. This morning I had a comprehensive hearing test for the first
time in well over a decade. I have a pronounced loss in higher
frequencies in one ear due to repeated ear infections as a small
child. One test was repeating single words spoken by the examiner. I
appears that where I had the most trouble was words with an
"unvoiced"(correct word?) initial consonant.

What would you suggest I read?
What would appropriate keyword(s) be for a Google search?
I'm retired. Therefore I've lots of time ;)

Thank you for your time.

Rafael Deliano

unread,
Sep 15, 2011, 2:49:00 PM9/15/11
to
> I don't know how to phrase the question.
> What would you suggest I read?

Hard to answer if the question isn´t too clear.

If one is technical/engineering oriented and likes old books:
Ainsworth "Mechanisms of Speech Recognition" Pergamon 1976
Often found cheap second hand at amazon:
http://www.amazon.com/Mechanisms-Recognition-International-natural-philosophy/dp/0080203949/ref=sr_1_1?ie=UTF8&qid=1316111461&sr=8-1
At 150 pages not too long, very readable ( un-academic,
lacking rigor and formulas with greek letters ).
Ainsworth did then experiments with speech synthesizers
to investigate human perception of slight variations in
speech that were aimed at improving speech recognition systems.

MfG JRD
0 new messages