Abstract

In this paper we describe a speaker-cluster
normalization algorithm that we applied to both gendernormalization
and speaker-normalization. To achieve
parameter sharing the acoustic space is partitioned into
classes. A maximum likelihood approach has been
proposed under which the delta between the
distribution mean and its corresponding acoustic class
is mostly speaker-independent, whereas the means of
the acoustic classes are mostly speaker-dependent.
When applied to gender-normalization, the error rate
reduction approaches that of a gender-dependent
system but with half the number of parameters. For a
speaker-normalized system, a 30% decrease in error
rate was obtained in a batch recognition experiment in
a context-dependent continuous-density HMM