Ex Parte Cutler et al

Patent Trial and Appeal Board

Jun 26, 2013

10968873 (P.T.A.B. Jun. 26, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte ANN R. CUTLER and ROBERT B. GREGORY
____________________

Appeal 2011-001804
Application 10/968,873
Technology Center 2600
____________________

Before CAROLYN D. THOMAS, JASON V. MORGAN, and
JUSTIN BUSCH, Administrative Patent Judges.

BUSCH, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2011-001804
Application 10/968,873

2
Appellants appeal under 35 U.S.C. § 134(a) from a final rejection of
claims 1-30, 32, and 33. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.

Introduction
According to Appellants, the invention relates to “methods and
systems for providing feedback speech instruction.” Spec. § Abstract.

STATEMENT OF THE CASE
Exemplary Claims1
Claims 1 and 13, reproduced below, are illustrative of the claimed
subject matter:
1. A method for providing feedback speech instructions
comprising the steps of:
(a) collecting data corresponding to a plurality of
prosodic expression parameters associated with a speaker, said
plurality of prosodic expression parameters correlating with
speaker efficacy, said collecting being performed by a speech
analysis device having an audio collection device and a storage
medium for storing the collected data,
(b) determining deviations of the collected data from an
ideal model of prosodic expression; and
(c) providing feedback to the speaker responsive to the
deviations.

1 Should this matter undergo further prosecution, the Examiner may wish to
review the claims to determine whether the claims recite statutory subject
matter under 35 U.S.C. § 101 or whether the claims are directed to mere
abstract mental processes. The Examiner also may wish to review the claims
to determine whether any construction of the term “ideal” would “depend
solely on the unrestrained, subjective opinion of a particular individual
purportedly practicing the invention.” See Datamize, LLC v. Plumtree
Software, Inc., 417. F.3d 1342, 1350 (Fed. Cir. 2005).
Appeal 2011-001804
Application 10/968,873

3
13. A method for developing a database of an ideal prosodic
speech model comprising the steps of:
(a) collecting ideal speech data corresponding to a
plurality of parameters associated with prosodic expressions of
at least one ideal speaker; wherein the plurality of parameters
comprises at least one of: pitch, volume, pitch variation,
volume variation, frequency of variation of pitch, frequency of
volume, rhythm, tone, speech cadence, vocal formants and the
change in vocal formants, and a combination thereof, said
plurality of parameters correlating with speaker efficacy; and
(b) determining an ideal prosodic speech model from the
collected ideal speech data by applying at least one pre-
determined algorithm.

References
Russell, et al.
Blass, et al.
Shrivastav
Handal, et al.
US 5,791,904
US 6,296,489 B1
US 2004/0167774 A1
US 6,963,841 B2
Aug. 11, 1998
Oct. 2, 2001
Aug. 26, 2004
Nov. 8, 2005

Rejections
Claims 1, 5-12, 20-27, 30, and 32 stand rejected under 35 U.S.C.
§ 102(b) as being anticipated by Blass. Ans. 4-5.
Claims 13-15, 17-19, and 33 stand rejected under 35 U.S.C. § 102(e)
as being anticipated by Handal. Ans. 5-7.
Claims 13 and 28 stand rejected under 35 U.S.C. § 102(e) as being
anticipated by Shrivastav. Ans. 7.
Claims 2-4 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Blass and Handal. Ans. 8-9.
Claim 16 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Handal and Russell. Ans. 9.
Appeal 2011-001804
Application 10/968,873

4
Claims 26 and 28 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Blass and Shrivastav. Ans. 9-10.

ANALYSIS
We have reviewed the Examiner’s rejections in light of Appellants’
arguments that the Examiner has erred. We disagree with Appellants’
conclusions.

Claim 1 Rejected as Anticipated by Blass
Appellants argue Blass does not anticipate claim 1 because Blass does
not teach speech analysis based on prosody and, therefore, does not teach
“collecting data corresponding to a plurality of prosodic expression
parameters associated with a speaker.” App. Br. 4-6. Specifically,
Appellants assert “prosody refers to qualities and characteristics of
connected speech as opposed to smaller elements like syllables or words.”
App. Br. 6. Appellants do not argue the Examiner’s findings of Blass’
disclosure, only that the Examiner’s construction is overly broad and that,
under Appellants’ proffered construction, claim 1 does not read on Blass.
App. Br. 4-6.
Issue 1: Did the Examiner err in finding Blass discloses “collecting
data corresponding to a plurality of prosodic expression parameters
associated with a speaker,” as recited in independent claim 1?

The Examiner finds Appellants admit that one definition of “prosody”
includes “‘the stress and intonation patterns of an utterance.’” Ans. 10
(quoting App. Br. 4-5). The Examiner further finds this definition to be the
Appeal 2011-001804
Application 10/968,873

5
most relevant definition because the Examiner’s proffered definition relates
to speech and linguistics, whereas the Appellants’ proffered definitions
relate to poetry, which is inconsistent with Appellants’ own use of
“prosody” in the Specification. Ans. 10-11. Moreover, regardless of
Appellants’ intended meaning, the Examiner finds the broadest reasonable
interpretation of Appellants’ claims consistent with Appellants’
Specification includes construing the term “prosodic” to mean relating to the
stress and intonation patterns of an utterance. Ans. 11-12. Under this
construction, the Examiner finds Blass meets the recited limitation because
Blass discloses extracting intonation from recorded speech, and comparing
the tone with a model. Ans. 10 (citing Blass 9:16-20).
We find no error in the Examiner’s construction of prosody and agree
the Examiner’s construction is a reasonable construction and consistent with
Appellants’ Specification. Moreover, even if we were to adopt Appellants’
construction, we note Appellants’ claim 1 merely recites “collecting data
corresponding to” prosodic expression parameters, the parameters
“correlating with” speaker efficacy. All speech corresponds to prosodic
expression parameters and some subset of those prosodic expression
parameters are correlated with speaker efficacy. Therefore, even under
Appellants’ construction, claim 1 would read on Blass because claim 1 does
not require any step of identifying or using prosodic expression parameters.
Rather, as discussed above, claim 1 merely requires that the recorded data
have a correlation with such parameters.
Given that we agree with the Examiner’s construction and that
Appellants’ arguments are limited to whether the Examiner’s construction is
unreasonable, we are unpersuaded by Appellants’ argument that Blass does
Appeal 2011-001804
Application 10/968,873

6
not disclose “collecting data corresponding to a plurality of prosodic
expression parameters associated with a speaker,” as recited in claim 1.
Thus, we sustain the Examiner’s rejection of claim 1 as being anticipated by
Blass.

Claim 13 Rejected as Anticipated by Handal
Appellants assert Handal relates to syllabic and word pronunciation
and, thus, does not teach “determining an ideal prosodic speech model . . .
by applying at least one pre-determined algorithm.” App. Br. 6. Moreover,
Appellants assert Handal does not even discuss applying an algorithm to
create a model. App. Br. 6.
Issue 2: Did the Examiner err in finding Handal discloses
“determining an ideal prosodic speech model . . . by applying at least one
pre-determined algorithm,” as recited in independent claim 13?

The Examiner finds Handal discloses receiving speech, extracting
tonal energy from the speech signal, and generating word and phrase models
for emotions. Ans. 12. The Examiner further finds Handal’s model is
developed from tonal energy, which is a prosodic model because prosody
includes speech intonation. Ans. 12 (citing Handal 9:46-10:9 and 15:53-
16:11). The Examiner also finds Handal creates its model via a step-by-step
method, which meets the broadest reasonable interpretation of “applying at
least one pre-determined algorithm.” Ans. 12-13 (citing Handal Fig. 1, 9:46-
11:20, which is a flowchart and accompanying description of a method for
creating a voice and error model in the “Development Phase”).
Appeal 2011-001804
Application 10/968,873

7
As discussed above with respect to claim 1, we find no error in the
Examiner’s construction of “prosody.” Thus, Appellants’ argument that
Handal relates to syllabic and word pronunciation is unavailing because our
construction of “collecting ideal speech data corresponding to a plurality of
parameters associated with prosodic expressions,” as recited in claim 13,
reads on Handal’s speech collection and tone extraction pointed to by the
Examiner. See Ans. 12. The Examiner’s construction of algorithm is
reasonable and Appellants’ have not provided sufficient argument or
evidence to persuade us that Handal does not disclose applying an algorithm.
Specifically, Handal’s method in the “Development Phase” discloses an
algorithm for generating a model. See Ans. 12-13.Thus, we sustain the
Examiner’s rejection of claim 13 as anticipated by Handal.

Claim 13 Rejected as Anticipated by Shrivastav
Appellants argue Shrivastav relates to processing voice and fails to
teach “extracting prosodic information therefrom.” App. Br. 7.
Issue 3: Did the Examiner err in finding Shrivastav discloses
“extracting prosodic information”?

The Examiner finds Shrivastav discloses extracting voice quality
attributes useful in measuring prosody, some of which are the same features
identified in Appellants’ claim 13. Ans. 13.
Appellants’ argument that Shrivastav doesn’t disclose “extracting
prosodic information” from collected speech is not commensurate with the
scope of the claim. Claim 13 merely recites collecting speech data
corresponding to parameters associated with prosodic expressions and
Appeal 2011-001804
Application 10/968,873

8
determining a speech model. Moreover, as discussed above, we agree with
the Examiner’s construction of “prosodic.” Further, the Examiner explains
on page 13 of the Answer that Shrivastav determines various voice
attributes, including some of the attributes specifically enumerated in a list
of parameters Appellants have identified in claim 13 as being associated
with prosodic expressions. Ans. 13 (“These voice quality attributes include
loudness, pitch (fundamental frequency), intensity, formant frequencies, et.
([0027]-[0029]) . . . these extracted features are some of the same features
specifically [e]numerated in [Appellant’s] claim.”).
Therefore, we sustain the Examiner’s rejection of claim 13 as being
anticipated by Shrivastav.

Remaining Rejections
Claims 2-12, 14-30, 32, and 33 were not argued separately and either
ultimately depend from or include substantially the same limitations as one
of claims 1 and 13. Therefore, we sustain the respective rejections for the
same reasons as discussed above.

DECISION
For the above reasons, the Examiner’s rejection of claims 1-30, 32,
and 33 is affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED
msc