Appeal No. 1999-2294
Application No. 08/807,430
With respect to claim 21, appellants argue that the claim requires a nonparametric
speech model including discrete training observations which represent a single utterance
by a single speaker as above with claim 19, and neither Journal nor Scott teaches or
suggests the claimed invention. (See brief at page 9.) We agree with appellants.
Furthermore, the examiner has not provided a convincing line of reasoning why it would
have been obvious to one of ordinary skill in the art at the time of the invention to apply a
nonparametric model as briefly mentioned in the Scott reference to the parametric model
as taught and suggested by the Journal reference.
The examiner’s only response to appellants’ arguments is to address the single
utterance by a single speaker. (See answer at page 7.) Clearly, the examiner’s position is
that the data at the time of acquisition are always single utterances by single speakers, but
this does not address the use of individual utterances by individual speakers in the
comparison rather than the use of a statistical model generated using the sampled
utterances. The examiner argues that a nonparametric vocabulary is not in the claims. Id.
We disagree with the examiner and agree with appellants that a nonparametric vocabulary
is expressly set forth in claims 19 and 21 and implicitly claimed in the language of claims 1
and 12. Since appellants have rebutted the examiner’s rejection, and the examiner has not
responded to appellants’ arguments, we accept appellants’ rebuttal. Therefore, we cannot
sustain the examiner’s rejection of claims 1, 12, 19, and 21 and their dependent claims.
7
Page: Previous 1 2 3 4 5 6 7 8 9 Next
Last modified: November 3, 2007