Ex Parte Weiner

Patent Trial and Appeal Board

May 8, 2018

13562733 (P.T.A.B. May. 8, 2018)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
13/562,733 07 /31/2012
86738 7590 05/10/2018
MCCARTER & ENGLISH, LLP BOSTON
265 Franklin Street
Boston, MA 02110
FIRST NAMED INVENTOR
Steven Weiner
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
ATTORNEY DOCKET NO. CONFIRMATION NO.
121439-00301 6420
EXAMINER
ORTIZ SANCHEZ, MICHAEL
ART UNIT PAPER NUMBER
2658
NOTIFICATION DATE DELIVERY MODE
05/10/2018 ELECTRONIC
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.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address( es):
docket@mccarter.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte STEVEN WEINER
Appeal2017-006620
Application 13/562,733 1
Technology Center 2600
Before ROBERT E. NAPPI, MARC S. HOFF, and
NORMAN H. BEAMER, Administrative Patent Judges.
BEAMER, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Appellant appeals under 35 U.S.C. § 134(a) from the Examiner's final
rejection of claims 1, 3-11, 13-21, and 23-30. Claims 2, 12, and 22 are
cancelled. We have jurisdiction over the pending rejected claims under
35 U.S.C. § 6(b).
We affirm.
1 Appellant identifies SRI International as the Real Party in Interest. (App.
Br. 1.)
Appeal2017-006620
Application 13/562,733
THE INVENTION
Appellant's disclosed and claimed invention is directed to
personalizing a voice user interface of a remote multi-user service, using a
language model specific to an identified user. (Abstract.) Independent
claim 1, reproduced below, is illustrative of the subject matter on appeal:
1. A computer-implemented method for personalizing a
voice user interface of a single remote multi-user service, the
method comprising:
providing a voice user interface for the single remote
multi-user service;
receiving voice information from an identified user at the
multi-user service through the voice user interface;
retrieving from memory, in at least one server of the
multi-user service, a language model specific to the identified
user and specific to the multi-user service, which models one or
more language elements relating to content stored by the multi-
user service and associated with a personal account of the
identified user, wherein the one or more language elements
comprise metadata that describes content stored by the remote
service in association with the personal account of the identified
user and that are associated with at least one prior instance of
voice information from the identified user;
applying the retrieved language model, with a processor,
to interpret the received voice information;
responding to the interpreted voice information; and
responsive to a change in the content maintained by the
multi-user service and associated with the user, updating by the
processor the language model specific to the identified user and
the multi-user service so as to reflect said change.
2
Appeal2017-006620
Application 13/562,733
REJECTION
The Examiner rejected claims 1, 3-11, 13-21, and 23-30 under
35 U.S.C. § 103(a) as being unpatentable over Van Kommer
(US 2007/0124134 Al, pub. May 31, 2007) (hereinafter "Kommer"), and
Zavaliagkos et al. (US 2013/0030804 Al, pub. Jan. 31, 2013) (hereinafter
"Zavaliagkos"). (Final Act. 3-18.)
ISSUE ON APPEAL
Appellant's arguments in the Appeal Brief present the following
issue: 2
Whether the Examiner erred in finding the combination of Kommer
and Zavaliagkos teaches or suggests the independent claim 1 limitation:
wherein the one or more language elements comprise metadata
that describes content stored by the remote service in
association with the personal account of the identified user and
that are associated with at least one prior instance of voice
information from the identified user ....
and the commensurate limitations of independent claims 11 and 21.
(Hereinafter, "the metadata limitation"). (App. Br. 9-22.)
ANALYSIS
We have reviewed the Examiner's rejections in light of Appellant's
arguments that the Examiner erred. We disagree with Appellant's
2 Rather than reiterate the arguments of Appellant and the positions of the
Examiner, we refer to the Appeal Brief (filed Sept. 15, 2016) (herein, "App.
Br."); the Reply Brief (filed Mar. 13, 2017) (herein, "Reply Br."); the Final
Office Action (mailed Mar. 18, 2016) (herein, "Final Act."); and the
Examiner's Answer (mailed Jan. 12, 2017) (herein, "Ans.") for the
respective details.
3
Appeal2017-006620
Application 13/562,733
arguments, and we adopt as our own ( 1) the pertinent findings and reasons
set forth by the Examiner in the Action from which this appeal is taken
(Final Act. 3-18) and (2) the corresponding findings and reasons set forth by
the Examiner in the Examiner's Answer in response to Appellant's Appeal
Brief. (Ans. 17-29.) We concur with the applicable conclusions reached by
the Examiner, and emphasize the following.
In finding that Kommer and Zavaliagkos teach or suggest the claimed
subject matter, the Examiner relies on the disclosure in Kommer of a
"multimodal business channel between users, service providers and network
operators," in which "user dependent language models" are generated by
speech recognition systems and are centrally stored in a database, as well as
on user devices. The language models are continuously adapted, through
user interactions and input, to improve the quality of speech recognition.
Users may verbally interact with the system to obtain services - for
example, to order travel packages, interact with call centers, or automatically
select music played to the user. The language models used for access to the
various services are updated on a daily basis, and the information used to
update the models includes users' metadata. (Final Act. 3-5; Kommer
Abstract, Fig. 3, i-fi-f 11,12, 16, 17, 40, 54, 70, 76-83, 91-102.)
The Examiner also relies on the disclosure in Zavaliagkos of a method
for improving the accuracy of transcription generated by an automatic
speech recognition engine, which uses personal data of a client, such as
metadata including the client's phone number or electronic address, and
web-based sources such as social networking or other service providers, to
improve the language models used by a central server, and send the
improved models back to the server to facilitate more accurate future
4
Appeal2017-006620
Application 13/562,733
transcriptions. (Final Act. 3-5; Zavaliagkos Abstract, Figs. 3, 4B, i-fi-f 17, 19,
23, 25, 31, 37, 41, 48, claim 24.)
Appellant argues the combination of Kommer and Zavaliagkos does
not teach or suggest the metadata limitation at issue. (App. Br. 12-15.)
Although the Specification provides no definition of "metadata" as used in
the claims, it discloses an example:
As one example, a user's media content can include music,
videos, and images, as well as metadata associated with the
music, videos, and images. Metadata for music can include, for
example, artist names, album titles, song titles, playlists, music
genres, and/or any other information related to the music.
Metadata for videos can include, for example, video titles (e.g.,
movie names), actor names, director names, movie genres,
and/or any other information related to the videos.
(Spec. i1 42.) The Specification further discloses the use of the metadata as
the basis for language models for speech recognition: "The personalized
language model can be constructed for the user based on user information
accessible by the user, such as, for example, the content of the user's multi-
user service account and/or the metadata associated therewith." (Spec.
i1 47 .)
In this context, we are not persuaded the Examiner erred in finding
that the combination of Kommer and Zavaliagkos teaches or suggests the
metadata limitation. In particular, for this limitation the Examiner relies on
the disclosure in Zavaliagkos of the use of a client's personal data- "such
as address book information, SMS and email messages, web-based sources,
etc."; or data "gathered from a social networking service or another service
provider" - as the basis for language models used for speech recognition.
(Final Act. 5-6; Ans. 19-22; Zavaliagkos i-fi-f 17, 19, 25, 34.) As the
5
Appeal2017-006620
Application 13/562,733
Examiner points out, this personal data includes metadata including a
"phone number or electronic address of a person." (Ans. 21; Zavaliagkos
claim 24, see also i-f 38 ("the term 'personal data' is intended to also include
all supplemental data the system uses to create the vocabulary or improve
the accuracy of the transcription").)
Appellant argues that "Zavagliagkos' personal data stored on a client
is simply not the 'content stored by the remote service in association with
the personal account of the identified user' recited in claim 1," because the
personal data of Zavaliagkos is stored on the client and not available to the
server. (App. Br. 14.) However, the claims do not require the personal data,
i.e., metadata, itself to be stored on a remote server - rather, the claims
require the "language model ... which models" the metadata to be stored in
a server memory. As the Examiner finds, Zavaliagkos discloses the speech
recognition engine may be "locally running on the personal device or
otherwise accessible by the personal device if stored/executed
elsewhere." (Ans. 23; Zavaliagkos i-f 33.) In particular, Zavaliagkos
discloses that after the client "rescores" the results of the speech recognition
engine, it "sends the results back to the server," which results include
updated language models of metadata encountered during speech
recognition. (Zavaliagkos i-f 17, see also Figs. 3 (elements 345, 350), 4B
(element 482), 6 (element 635), i-fi-131, 37, 41, 48, 72, 74.) The results are
sent back to the server in order "to be considered during future
transcriptions." (Zavaliagkos i-f 41.)
Appellant further argues the Examiner errs in combining the
references because "Zavaliagkos and Kommer teach away from each other
in that they stem from opposite paradigms." (App. Br. 16.) Appellant
6
Appeal2017-006620
Application 13/562,733
argues Zavaliagkos is devoted to improving transcription accuracy and is
premised on users protecting privacy and avoiding sharing personal
information, while Kommer does not mention transcription, and is premised
on users giving up privacy for particular benefits. (Id.) These arguments are
unpersuasive. As discussed above, both references are directed to
improvements in automatic speech recognition, and in particular to the use
of language models personalized in accord with individual user information.
Also as discussed above, although Zavaliagkos confines user information to
the local users devices, it shares the language models based on the user
information with the central server, similar to the approach of Kommer and
the preferred embodiment of Appellant's Specification. Moreover, "[t]he
test for obviousness is not whether the features of a secondary reference may
be bodily incorporated into the structure of the primary reference ....
Rather, the test is what the combined teachings of those references would
have suggested to those of ordinary skill in the art." In re Keller, 642 F.2d
413, 425 (CCPA 1981). "Combining the teachings of references does not
involve an ability to combine their specific structures." In re Nievelt, 482
F.2d 965, 968 (CCPA 1973).
We therefore are not persuaded that the Examiner erred in combining
Zavaliagkos and Kommer based on the motivation to improve the
capabilities of Kommer using the teachings of Zavaliagkos "for the benefit
of improving the transcription[] accuracy by rescoring using the
personalized data." (Final Act. 6; see also Ans. 26-27.) As the Examiner
further finds, and we agree:
In regards to the motivation to combine both references
are within the same field of endeavor Kommer teaches methods
for personalizing services in a telecommunications network
7
Appeal2017-006620
Application 13/562,733
which involves generating user dependent language models in a
speech recognition system and Zavaliagkos teaches methods for
improving accuracy of a transcription in an automatic speech
recognition engine. The Zavaliagkos invention would improve
on the Kommer speech recognizer by using a personal profile of
a user which is what is taught in the claims.
(Ans. 19.)
Appellant does not point to any evidence of record that the resulting
combination would be "uniquely challenging or difficult for one of ordinary
skill in the art" or "represented an unobvious step over the prior art."
Leapfrog Enters., Inc. v. Fisher-Price, Inc., 485 F.3d 1157, 1162 (Fed. Cir.
2007) (citing KSR Int'! Co. v. Teleflex Inc., 550 U.S. 398, 418-19 (2007)).
The Examiner's findings are reasonable because the skilled artisan would
"be able to fit the teachings of multiple patents together like pieces of a
puzzle" because the skilled artisan is "a person of ordinary creativity, not an
automaton." KSR, 550 U.S. at 420-21. We are persuaded the claimed
subject matter exemplifies the principle that "[ t ]he combination of familiar
elements according to known methods is likely to be obvious when it does
no more than yield predictable results." KSR, 550 U.S. at 416.
Accordingly, we sustain the Examiner's obviousness rejection of
independent claims 1, 11, and 21.
8
Appeal2017-006620
Application 13/562,733
CONCLUSION
For the reasons stated above, we sustain the Examiner's obviousness
rejection of independent claims 1, 11, and 21 over Kommer and
Zavaliagkos. We also sustain the obviousness rejections of claims 3-10, 13-
20, and 23-30 over Kommer and Zavaliagkos, which rejections are not
argued separately with particularity. (App. Br. 17, 19, 21-22.)
DECISION
We affirm the Examiner's decision rejecting claims 1, 3-11, 13-21,
and 23-30.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l )(iv). See
37 C.F.R. § 41.50(f).
AFFIRMED
9