Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!howland.reston.ans.net!news.sprintlink.net!EU.net!ieunet!curia!csws5.ucc.ie!turlough
From: turlough@odyssey.ucc.ie (Turlough FitzPatrick)
Subject: RE:McGurk Effect
Message-ID: <D4t8s7.E6H@curia.ucc.ie>
Sender: usenet@curia.ucc.ie
Reply-To: turlough@odyssey.ucc.ie
Organization: Computer Science Dept., University College, Cork, Ireland
Date: Thu, 2 Mar 1995 11:19:18 GMT
Lines: 12

You might wounder do children use purely
audio input when first learning to talk or
do they use both audio and visual "input" stimuli
(probably more likely). If then people were intent
on modeling human hearing(audio perception)
surely they should incorporate visual input. Current
speech recognition systems provide high recognition
but I dont think they can get 99-100% acuracy. Maybe
the additional information contained in the speakers
facial expression is what is lacking to complete 
the picture.

