Ex Parte Huboi

Board of Patent Appeals and Interferences

Aug 18, 2011

09459380 (B.P.A.I. Aug. 18, 2011)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte PETER ALLEN HUBOI
___________

Appeal 2011-001837
Application 09/459,380
Technology Center 2600
____________

Before ERIC S. FRAHM, KRISTEN L. DROESCH, and ERIC B. CHEN,
Administrative Patent Judges.

CHEN, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2011-001837
Application 09/459,380

2
This is an appeal under 35 U.S.C. § 134(a) from the final rejection of
claims 30-46, 49, 50, 53 and 54.1 Claims 4, 13, 20 and 26 have been
cancelled. We have jurisdiction under 35 U.S.C. § 6(b). We affirm.
Appellant’s invention relates to methods and systems for analyzing
voice information to determine if specific words or phrases are used or if the
voice information exhibits a particular pattern of speech. (Spec. Abstract.)
If specified words or phrases are determined to have been used, a specific
action is taken. (Spec. Abstract.)
Claim 30 is exemplary, with disputed limitations in italics:
30. A method for processing a voice message, comprising:
storing one or more actions;
receiving a voice message;
analyzing the voice message to determine if the voice message
exhibits a predetermined pattern of speech, the predetermined pattern
of speech representing at least one of a tone of speech in the voice
message and a frequency of the speech in the voice message; and
performing one or more of the stored actions, if the
predetermined pattern of speech is found to occur in the voice
message.
Claims 30, 32, 35-37, 39-41, 43, 44, 46, 49 and 50 stand rejected
under 35 U.S.C. § 103(a) as being obvious over McDonough (U.S. Patent
No. 5,625,748) and Furui (Sadaoki Furui, DIGITAL SPEECH PROCESSING,
SYNTHESIS, AND RECOGNITION 225-289 (Marcel Dekker, Inc., 1989)).2

1 The rejections of claims 1-3, 5-12, 14-19, 21-25, 27-29, 47, 48, 51 and 52
under 35 U.S.C. § 103(a) have been withdrawn by the Examiner. (Ans. 2.)
2 Both Appellant and the Examiner have erroneously included Epstein in the
statement of the rejection. (Br. 20, 51; Ans. 4.)
Appeal 2011-001837
Application 09/459,380

3
Claims 31, 33, 34, 38, 42, 45, 53 and 54 stand rejected under 35
U.S.C. § 103(a) as being obvious over McDonough, Furui and Epstein (U.S.
Patent No. 6,327,343).
We are not persuaded by Appellant’s arguments (Br. 51-53) that the
combination of McDonough and Furui would not have rendered obvious
independent claim 30, including the disputed limitation “the predetermined
pattern of speech representing at least one of a tone of speech in the voice
message and a frequency of the speech in the voice message.”
The Examiner acknowledged that McDonough does not teach this
disputed claim limitation (Ans. 5) and cited Furui for the disclosure of
automatic speech recognition methods with “a lattice taking account of
allophones, coarticulation, stress, and syllables” corresponding to the
claimed “tone of speech” (Ans. 5-6; Furui, p. 276, ll. 1-15). The Examiner
also cited Furui for the disclosure of automatic speech recognition methods
with “Markov models for recognition of input speech converted into spectral
feature vectors by DFT [discrete Fourier transform]” corresponding to the
claimed “frequency of the speech.” (Ans. 6; Furui, p. 278, ll. 3-9.) The
Examiner concluded that independent claim 30 would have been obvious
over the combination of McDonough and Furui. (Ans. 6.)
Appellant argues that “the Examiner’s interpretation is unreasonably
broad and completely ignores the Specification” (Br. 51) because “the terms
‘tone of speech’ and ‘frequency of speech’ in Claim 30 means non-linguistic
characteristics, such as high pitched, loud, breathless, or rapid speech”
(Br. 52). However, the scope of the claim is not so narrow. The portions of
the Specification cited by Appellant do not provide adequate support for
requiring a “tone of speech” and “frequency of speech” to mean non-
Appeal 2011-001837
Application 09/459,380

4
linguistic characteristics. Appellant’s Specification describes that in one
embodiment, “rather than analyzing the voice information for specific words
or phrases, the voice information is analyzed looking for particular speech
characteristics, such as frequency and tone.” (Spec. 9:29-31; see also
Spec. 7:7-8.) The Specification further states that “[f]or example, a
prospective customer’s speech may be analyzed to determine if they are
angry by analyzing their speech for characteristics indicative of a person
being angry.” (Spec. 10:1-3.) As another example for an urgent call, “a
caller may speak rapidly, be out of breath, or be speaking in a high pitch.”
(Spec. 7:9-10.) In other words, the Specification discloses that analyzing a
person’s speech for characteristics of anger or urgency are only examples of
“frequency and tone” of speech.
The broadest reasonable meaning of “tone” includes the relevant plain
meaning which is “any sound considered with reference to its quality, pitch,
strength, source, etc.” Random House Dictionary of the English Language
1994 (2nd ed. 1987). Thus, we agree with the Examiner (Ans. 6) that Furui
teaches the claimed “tone of speech” because Furui describes analysis of
speech on the basis of allophones, coarticulation, stress and syllables (p. 276,
ll. 11-15).
The broadest reasonable meaning of “frequency” includes the relevant
plain meaning which is “rate of occurrence.” Random House Dictionary of
the English Language 766 (2nd ed. 1987). We agree with the Examiner
(Ans. 6) that Furui teaches the “frequency of speech” limitation based on
Furui’s description of using Markov models for recognition of input speech
converted into spectral feature vectors by DFT. One with ordinary skill in
the art would understand that the conversion of input speech into spectral
Appeal 2011-001837
Application 09/459,380

5
feature vectors (or spectral analysis) is based on the sound wave frequencies
of the input speech (i.e., the rate of occurrence of the sound waves).
Furthermore, McDonough also meets the limitation because McDonough
describes a speech event frequency detector 12 (col. 5, ll. 45-46) such that
“[t]he event frequencies are preferably estimates of the frequency of
occurrence of the potential speech events in the spoken data” (col. 5, ll. 54-
56; col. 7, ll. 26-28).
Therefore, we agree with the Examiner that the combination of
McDonough and Furui would have rendered obvious independent claim 30,
including the disputed limitation “the predetermined pattern of speech
representing at least one of a tone of speech in the voice message and a
frequency of the speech in the voice message.”
Accordingly, we sustain the rejection of independent claim 30 under
35 U.S.C. § 103(a). Claims 35 and 36 depend from independent claim 30
and Appellant has not presented any substantive arguments with respect to
these claims. Therefore, we sustain the rejection of these claims under 35
U.S.C. § 103(a) for the reasons discussed with respect to independent
claim 30.
Independent claims 37, 41, 44, 49 and 50 recite limitations similar to
those discussed with respect to independent claim 30 and Appellant has not
presented any substantive arguments with respect to these claims. We
sustain the rejection of claims 37, 41, 44, 49 and 50, as well as claims 40, 43
and 46, which depend from claims 37, 41 and 44, for the reasons discussed
with respect to claim 30.
We are not persuaded by Appellant’s arguments (Br. 54) that the
combination of McDonough and Furui would not have rendered obvious
Appeal 2011-001837
Application 09/459,380

6
dependent claims 32 and 39, including the disputed limitations “the user
specifying one or more actions” and “storing the user specified one or more
actions.”
The Examiner found that the system of McDonough that queries a
user as to the correctness of a proposed action corresponds to the limitation
of a “user specifying one or more actions.” (Ans. 7; McDonough, col. 2,
ll. 19-24.) The Examiner also found that the system of McDonough which
adds new words to its vocabulary corresponds to the limitation “storing the
user specified one or more actions.” (Ans. 7; McDonough, col. 2, ll. 12-15.)
We agree with the Examiner.
McDonough describes “speech recognition systems, and more
particularly . . . an improved topic discriminator of a spoken message.”
(Col. 1, ll. 7-9.) In the “Background of the Invention” section, McDonough
describes “[a] topic discriminator for spoken data . . . used to classify the
data into one of a set of known topics” (col. 1, ll. 12-13) for example, a
system for “routing of messages to one of several departments of a retail
store” (col. 2, ll. 3-4). McDonough also teaches that “an unknown word can
be identified [by the system] in an incoming message” and that “[t]he new
word is then added to the vocabulary.” (Col. 2, ll. 12-14.) Furthermore,
“[t]his learning process . . . requires that the system be able to query the user
as to the correctness of the action it proposes (e.g., ‘Would you like to be
connected with the furniture department?’), and subsequently re-learn those
messages which produce undesirable recommendations.” (Col. 2, ll. 19-24.)
In other words, McDonough teaches “the user specifying one or more
actions” and “storing the user specified one or more actions.”
Appeal 2011-001837
Application 09/459,380

7
Appellant argues that McDonough does not teach the disputed
limitations because “McDonough describes a system in which ‘it is assumed
that all possible actions are known,’ while ‘the system has the capacity to
interactively learn new vocabulary words’ that a user may associate with a
desired action” and thus, “all possible actions are already specified and
stored in the system, new vocabulary words can be learned, and a user can
associate the new words with an existing action.” (Br. 54.) However, the
arguments presented do not provide an adequate explanation as to why
McDonough does not teach the disputed limitations.
Therefore, we agree with the Examiner that the combination of
McDonough and Furui would have rendered obvious dependent claims 32
and 39, including the disputed limitations “the user specifying one or more
actions” and “storing the user specified one or more actions.”
Accordingly, we sustain the rejection of dependent claims 32 and 39
under 35 U.S.C. § 103(a).
We are not persuaded by Appellant’s arguments (Br. 45-46) that the
combination of McDonough, Furui and Epstein would not have rendered
obvious dependent claim 33, including the disputed limitation “wherein the
stored actions include marking the message as urgent.”
The Examiner acknowledged that McDonough does not teach this
disputed claim limitation (Ans. 10) and cited Epstein for the disclosure of a
system 10 including an audio indexer/prioritizer module 34 for retrieving
urgent messages (Ans. 11, 17; Epstein, col. 13, ll. 57-62). The Examiner
concluded that dependent claim 33 would have been obvious over the
combination of McDonough, Furui and Epstein. (Ans. 11.) We agree with
the Examiner.
Appeal 2011-001837
Application 09/459,380

8
As discussed previously, McDonough describes “speech recognition
systems, and more particularly . . . an improved topic discriminator of a
spoken message.” (Col. 1, ll. 7-9.) Epstein describes “[a] programmable
automatic call and data transfer processing system which automatically
processes incoming telephone calls, facsimiles and e-mails based on the
identity of the caller or author, the subject matter of the message or request,
and/or the time of day.” (Abstract.) In one embodiment, a system 10 that
automatically answers an incoming telephone call (col. 3, ll. 34-35) includes
an audio indexer/prioritizer module 34 programmed to index messages and
conversations, for example, to “retrieve all urgent messages” (col. 13, ll. 57-
62).
A person of ordinary skill in the art at the time of the invention would
have recognized that incorporating the known audio indexer/prioritizer
module 34 of Epstein with the speech recognition systems of McDonough
would improve McDonough by providing the ability to retrieve urgent
messages. See KSR Int’l Co. v. Teleflex, Inc., 550 U.S. 398, 417 (2007).
Thus, we agree with the Examiner (Ans. 11) that modifying McDonough to
include the audio indexer/prioritizer module 34 of Epstein would have been
obvious.
Appellant argues that “[n]owhere in the description of handling a
voice call does Epstein describe storing urgency information relating to the
call, as is described for facsimile and e-mail messages.” (Br. 46.) Contrary
to Appellant’s argument and as discussed previously, Epstein describes a
programmable automatic call and data transfer processing system (Abstract)
that includes an audio indexer/prioritizer module 34 programmed to retrieve
all urgent messages (col. 13, ll. 57-62).
Appeal 2011-001837
Application 09/459,380

9
Therefore, we agree with the Examiner that the combination of
McDonough, Furui and Epstein would have rendered obvious dependent
claim 33, including the disputed limitation “wherein the stored actions
include marking the message as urgent.”
Accordingly, we sustain the rejection of dependent claim 33 under 35
U.S.C. § 103(a).
We are not persuaded by Appellant’s arguments (Br. 47) that the
combination of McDonough, Furui and Epstein would not have rendered
obvious dependent claim 34, including the disputed limitation “wherein the
stored actions include calling a pager.”
The Examiner acknowledged that McDonough does not teach this
disputed claim limitation and cited Epstein for the disclosure of a system 10
that transforms an incoming telephone call into a page. (Ans. 11; Epstein,
col. 4, ll. 1-3.) The Examiner concluded that dependent claim 34 would
have been obvious over the combination of McDonough, Furui and Epstein.
(Ans. 11-12.) We agree with the Examiner.
As discussed previously, McDonough describes “speech recognition
systems, and more particularly . . . an improved topic discriminator of a
spoken message.” (Col. 1, ll. 7-9.) Epstein describes a system 10 that
automatically answers an incoming telephone call (col. 3, ll. 34-35) such that
the system 10 is “programmed to transform an incoming telephone call or
messages into a page which can then be transmitted to the user’s pager”
(col. 4, ll. 1-3) when the user is not at home or away from the office (col. 14,
ll. 46-49).
A person of ordinary skill in the art at the time of the invention would
have recognized that incorporating the known method of transferring an
Appeal 2011-001837
Application 09/459,380

10
incoming call into a page, as taught by Epstein, with the speech recognition
systems of McDonough would improve McDonough by providing the
ability to transmit a message to a user who is not at home or out of the
office. See KSR, 550 U.S. at 417. Thus, we agree with the Examiner
(Ans. 11-12) that modifying McDonough to include Epstein’s method of
transferring an incoming call into a page would have been obvious.
Therefore, we agree with the Examiner that the combination of
McDonough, Furui and Epstein would have rendered obvious dependent
claim 34, including the disputed limitation “wherein the stored actions
include calling a pager.”
Accordingly, we sustain the rejection of dependent claim 34 under 35
U.S.C. § 103(a).
Although Appellant nominally argues the rejection of claims 31, 38,
42, 45, 53 and 54 separately (Br. 44-45, 48-50), the arguments presented do
not point out with particularity or explain why the limitations of the claims
are separately patentable. Accordingly, we sustain the rejection of
claims 31, 38, 42, 45, 53 and 54 under 35 U.S.C. § 103(a).

DECISION
The decision to reject claims 30-46, 49, 50, 53 and 54 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