Ex Parte Budde et al

Board of Patent Appeals and Interferences

Jul 2, 2010

10276071 (B.P.A.I. Jul. 2, 2010)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
10/276,071 11/12/2002 Mark Budde 2177.1005 9402
21171 7590 07/06/2010
STAAS & HALSEY LLP
SUITE 700
1201 NEW YORK AVENUE, N.W.
WASHINGTON, DC 20005
EXAMINER
HAN, QI
ART UNIT PAPER NUMBER
2626
MAIL DATE DELIVERY MODE
07/06/2010 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte MARK BUDDE and TOBIAS SCHNEIDER
____________

Appeal 2009-012222
Application 10/276,071
Technology Center 2600
____________

Before ROBERT E. NAPPI, JOHN C. MARTIN, and
JOSEPH F. RUGGIERO, Administrative Patent Judges.

RUGGIERO, Administrative Patent Judge.

DECISION ON APPEAL1

1 The two-month time period for filing an appeal or commencing a civil
action, as recited in 37 C.F.R. § 1.304, or for filing a request for rehearing,
as recited in 37 C.F.R. § 41.52, begins to run from the “MAIL DATE”
(paper delivery mode) or the “NOTIFICATION DATE” (electronic delivery
mode) shown on the PTOL-90A cover letter attached to this decision.
Appeal 2009-012222
Application 10/276,071

2
STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134 from the Final Rejection of
claims 12-32, which are all of the pending claims. Claims 1-11 have been
canceled. An oral hearing was conducted on this appeal on June 23, 2010.
We have jurisdiction under 35 U.S.C. § 6(b).
We reverse.
Rather than reiterate the arguments of Appellants and the Examiner,
we refer to the Brief (filed July 8, 2008) and the Answer (mailed September
26, 2008) for the respective details. We have considered in this decision
only those arguments Appellants actually raised in the Brief. Any other
arguments which Appellants could have made but chose not to make in the
Brief are deemed to be waived. See 37 C.F.R. § 41.37(c)(1)(vii).

Appellants’ Invention
Appellants’ invention relates to the creation of a speech database for a
target vocabulary for training a speech recognition system. The words of the
target vocabulary are composed of segments which have one or more
phonemes and are derived from a generic spoken training text which is
independent of the target vocabulary. (See generally Spec. ¶¶ [0008]-
[0009]).
Claim 12 further illustrates the invention and reads as follows:
12. A method for creating a speech database for training of a
speech recognition system with a target vocabulary, comprising:
converting the words of the target vocabulary into a phonetic
description so that the individual words are represented by a sequence of
phonemes; and
concatenating segments of a generic spoken training text to form the
words of the target vocabulary and thereby the speech database, each
Appeal 2009-012222
Application 10/276,071

3
segment containing at least one phone, the segments being concatenated to
form a sequence of phones which correspond respectively to the sequence of
phonemes of the target vocabulary.

The Examiner’s Rejections
The Examiner’s Answer cites the following prior art references:
Gould US 5,850,627 Dec. 15, 1998
Sabourin US 6,108,627 Aug. 22, 2000
(filed Oct. 31, 1997)
Miller US 6,134,528 Oct. 17, 2000
(filed Jun. 13, 1997)
Bahl US 6,260,014 B1 Jul. 10, 2001
(filed Sep. 14, 1998)
Appellants’ admitted prior art (AAPA) appearing at paragraph 4 of the
Specification.
Claims 12, 13, 16, 19, 28, and 31 stand rejected under 35 U.S.C. §
103(a) as being unpatentable over the combination of Bahl and Miller.
Claims 14, 15, and 22 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over the combination of Bahl, Miller, and Sabourin.
Claims 17, 18, and 21 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over the combination of Bahl, Miller, and “well known
prior art.”
Claim 20 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Bahl, Miller, and Gould.
Claims 23-25 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Bahl, Miller, Sabourin, and “well
known prior art.”
Claims 26-30 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Bahl, Miller, Sabourin, and Gould.
Appeal 2009-012222
Application 10/276,071

4
Claim 32 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Bahl, Miller, and AAPA.

ISSUES
Based on Appellants’ contentions, as well as the findings and
conclusions of the Examiner, the pivotal issues before us are whether the
Examiner erred in determining:
a) the obviousness to the ordinarily skilled artisan of combining
the teachings of Bahl, Miller, and various secondary references; and
b) if obvious to combine, whether the resulting combination
satisfies the requirements of appealed claims 12-32.

PRINCIPLES OF LAW
In rejecting claims under 35 U.S.C. § 103, it is incumbent upon the
Examiner to establish a factual basis to support the legal conclusion of
obviousness. See In re Fine, 837 F.2d 1071, 1073 (Fed. Cir. 1988). In so
doing, the Examiner must make the factual determinations set forth in
Graham v. John Deere Co., 383 U.S. 1, 17 (1966) (stating that 35 U.S.C.
§ 103 leads to three basic factual inquiries: the scope and content of the prior
art, the differences between the prior art and the claims at issue, and the
level of ordinary skill in the art). Furthermore,
“there must be some articulated reasoning with some rational
underpinning to support the legal conclusion of obviousness” . .
. . [H]owever, the analysis need not seek out precise teachings
directed to the specific subject matter of the challenged claim,
for a court can take account of the inferences and creative steps
that a person of ordinary skill in the art would employ.
Appeal 2009-012222
Application 10/276,071

5
KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007) (quoting In re Kahn,
441 F.3d 977, 988 (Fed. Cir. 2006)).

ANALYSIS
I. The obviousness rejection of claims 12, 13, 16, 19, 28,
and 31 based on the combination of Bahl and Miller.
With respect to the Examiner’s obviousness rejection of appealed
independent claims 12 and 31, Appellants’ arguments focus (App. Br. 7-12)
on the alleged deficiency of Bahl in disclosing the claimed feature of
“concatenating segments of a generic spoken training text to form the words
of the target vocabulary.” This concatenation feature is further recited as
forming “a sequence of phones which correspond respectively to the
sequence of phonemes of the target vocabulary.” According to Appellants
(App. Br. 14-16), Bahl has no disclosure of concatenating segments of a
spoken generic training text to form the words of a target vocabulary.
We agree with Appellants. In addressing the language of independent
claims 12 and 32, the Examiner (Ans. 4, 14) has initially directed attention
to the portion of the disclosure of Bahl at column 3, lines 29-53, which
describes the construction of an acoustic model which is to be used for
speech recognition. This acoustic model construction is disclosed as
involving the extraction of feature vectors from input speech and the
alignment of the feature vectors with stored phone sequences which
correspond to the uttered input speech. We find nothing in this portion of
Bahl’s disclosure which teaches or suggests the concatenation of segments
from a generic spoken text to form words of a target vocabulary as claimed.
We also find to be deficient the Examiner’s reliance (Ans. 4, 14) on
the portion of the disclosure at column 3, line 54, through column 4, line 49,
Appeal 2009-012222
Application 10/276,071

6
of Bahl, which includes a description of phonetic baseform Tables 1 and 2.
This portion of Bahl describes an example of a speech recognition
application having particular interest in recognizing speech utterances of
digits zero through nine. Table 1 summarizes the phonetic baseforms for
digits zero through nine developed as part of a generic recognizer. This digit
subset of the generic recognizer is then used to construct a task specific
model by adding new acoustic units to the phonetic baseforms of Table 1 to
develop new phonetic baseforms as illustrated in Table 2.
According to Bahl (col. 5, ll. 49-52), the composite generic and task
specific acoustic models are effective in dealing with large vocabulary
applications while improving the recognition performance of a subset of the
vocabulary application that might be of particular importance such as
recognizing digits. Again, however, we agree with Appellants that there is
no teaching or suggestion in Bahl’s disclosure of the development of a task
specific model of concatenating segments of a generic spoken text to form
words of a target vocabulary. We also note that Appellants’ claimed
concatenating feature also requires the formation of “a sequence of phones
which correspond respectively to the sequence of phonemes of the target
vocabulary,” a feature which we find lacking in the disclosure of Bahl.
Lastly, we agree with Appellants that Miller, relied upon by the
Examiner (Ans. 4, 5, 15, 16) as providing an express teaching of creating a
speech database, does not overcome the above discussed deficiencies of
Bahl. As argued by Appellants (App. Br. 15-16), the text-to-speech
synthesis system of Miller is directed (Fig. 4; col. 5, ll. 28-32) to adapting an
already existing speech database to account for how words are used in
context. Miller describes the segmentation of input speech in which the ends
Appeal 2009-012222
Application 10/276,071

7
of words are associated with a particular time in the course of the utterance.
As such, the labeled speech database in Miller, referenced by the Examiner,
is only used for adapting lexical pronunciations in an already existing
dictionary database. We find, as argued by Appellants, no teaching or
suggestion of concatenating segments of text to form words of a target
vocabulary and, particularly, no disclosure of concatenating segments from a
generic spoken training text.
With the above discussion in mind, it is apparent to us that even
assuming, arguendo, that the Examiner provided a proper basis for
combining the disclosures of Bahl and Miller, the resulting combination
would not satisfy the claimed features present in each of the appealed
independent claims 12 and 31. We, therefore, do not sustain the Examiner’s
35 U.S.C. § 103(a) rejection of independent claims 12 and 31, nor of claims
13, 16, 19, and 28 dependent thereon.

II. The obviousness rejections of independent claim 32 and
dependent claims 14, 15, 17, 18, 20-27, 29, and 30.
We also do not sustain the Examiner’s obviousness rejection of
independent claim 32 and dependent claims 14, 15, 17, 18, 20-27, 29, and 30
in which the various cited secondary references are applied in various
combinations with the combination of Bahl and Miller. We find nothing in
these secondary references which overcomes the innate deficiencies of Bahl
and Miller as discussed supra.

Appeal 2009-012222
Application 10/276,071

8
CONCLUSION OF LAW
Based on the findings of facts and analysis above, we conclude that
Appellants have shown the Examiner erred in rejecting claims 12-32 for
obviousness under 35 U.S.C. § 103(a).

DECISION
The Examiner’s 35 U.S.C. § 103(a) rejection of claims 12-32, all of
the appealed claims, is reversed.

REVERSED

babc

STAAS & HALSEY LLP
SUITE 700
1201 NEW YORK AVENUE, N.W.
WASHINGTON, DC 20005