Appeal 2007-3257
Application 10/046,632
estimation unit in the Specification (FF 3), the sinusoidal coder of Kleijn
teaches the claimed sinusoidal estimation.
We remain unconvinced by Appellant’s argument that the applied
prior art must also show that the transform is an algorithmic estimation, or
that the resulting spectrum is an estimation (Br. 9). We note that the claims
neither recite an algorithmic estimation nor include limitations that require
an “algorithmic transformation.” Additionally, we disagree with Appellant
that the taking of a fast Fourier Transform of Kleijn relates to sinusoidal
coding and not the claimed estimating sinusoidal code (Br. 10), since
Appellant’s own disclosure provides for the same fast Fourier Transform as
the way of carrying out the estimation (FF 3). Therefore, as broadly as
claimed and consistent with the Specification, Appellant’s recited sinusoidal
estimation unit reads on the coder of Kleijn. In other words, one of ordinary
skill in the art would have used the encoder of Elder in combination with the
sinusoidal estimation unit taught by Kleijn for benefiting from its natural
processing of the spectrum for voiced speech (FF 1).1
On the record before us, it follows that in this case Appellant has not
shown that the Examiner erred in rejecting Claim 1 under the § 103(a)
rejection. Since Appellant’s arguments focus on the patentability of claim 1
without addressing the various rejections of the remaining claims (Br. 10-
11), claims 2-17 fall with claim 1.
1 Although not discussed by the Examiner, we note that the background of
the invention in Figure 5 also discloses a sinusoidal estimation unit 140’ as a
known component in a parametric encoder (Specification 1).
6
Page: Previous 1 2 3 4 5 6 7 Next
Last modified: September 9, 2013