Appeal No. 2002-2207
Application 09/057,729
The disclosed invention pertains to a method and
apparatus for use in the field of speech recognition.
Representative claim 1 is reproduced as follows:
1. A method for recognising speech, wherein a recognisable
speech signal is divided in time into successive frames of
specific length, each speech frame is analysed for producing at
least one parameter per frame, illustrating the speech signal,
said parameters, relating to each frame, are stored in a sliding
buffer for minimizing the delay due to the normalization process
for calculation of normalisation coefficients for each frame,
said parameters are modified utilising said normlisation
coefficients and speech recognition is carried out utilising the
modified parameters, wherein only part of the successive
parameters are stored periodically and at least one parameter is
modified on the basis of the parameters stored periodically in
order to produce said modified parameter, and for said
modification, a standard deviation of said periodically stored
parameters is defined, wherein only part of the stored parameters
are used at the beginning of the speech recognition.
The examiner relies on the following references:
Satoh et al. (Satoh) 5,293,588 Mar. 08, 1994
Kroeker et al. (Kroeker) 5,369,726 Nov. 29, 1994
Claims 1-14 stand rejected under 35 U.S.C. § 103. As
evidence of obviousness the examiner offers Satoh in view of
Kroeker.
Rather than repeat the arguments of appellants or the
examiner, we make reference to the briefs and the answer for the
respective details thereof.
-2-
Page: Previous 1 2 3 4 5 6 7 8 Next
Last modified: November 3, 2007