Appeal No. 1996-2198
Application 08/077,506
block 8 generates probability distributions A and B, which
Kuroda represents as parameter P(i,j,k). These discrete
parameter values, which are stored in parameter table 11 in
the manner shown in Figure 6, represent the probability that a
transition from the state i to the state j occurs in a Markov
model and that a label k is produced at that i6j transition
(col. 4, lines 33-40). For the foregoing reasons, the
examiner's additional reliance on adaptation block 9, which
tailors the stored parameter values to a different speaker or
to the same speaker at a different time (col. 1, lines 55-68;
col. 7, lines 12-45), is unnecessary.
The examiner next relies on Kuroda's training and
adaptation blocks 8 and 9 in combination with Baker to satisfy
claim 1's requirement for "function generating means . . . for
generating an approximate continuous probability density
function." Specifically, the examiner states that such a
means is
suggested by Kuroda's training [block] 8 and
adaptation [block] 9, fig. 1[,] for calculating new
parameters by computing a weighted sum of the
probabilistic frequencies and Baker's figure 2[,]
which is a schematic representation of how phonetic
-17-
Page: Previous 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Next
Last modified: November 3, 2007