Bimodal Recognition of Italian Plosives

A bimodal automatic speech
recognition system, in which the speech
signal is synchronously analyzed by an
audio channel producing spectral-like
parameters every 2 ms and by a visual
channel computing lip and jaw kinematic
parameters, is described and some results
are given for various speaker independent
phonetic recognition experiments
regarding the Italian plosive class in
different noisy conditions.

Publication type: 
Contributo in atti di convegno
Author or Creator: 
Cosi P.
Dugatto M.
Ferrero F.
Magno Caldognetto E.
Vagges K.
XIII International Congress of Phonetic Sciences ICPhS-95, pp. 260–263, Stochkolm, Sweden, 14-18 August, 1995
Resource Identifier: