Ex Parte Ullrich et al

Board of Patent Appeals and Interferences

Mar 30, 2009

10960775 (B.P.A.I. Mar. 30, 2009)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte MEINHARD ULLRICH, ERIC THELEN,
and STEFAN BESLING
_____________

Appeal 2009-1561
Application 10/960,775
Technology Center 2600
____________

Decided:1 March 30, 2009
____________

Before MAHSHID D. SAADAT, THOMAS S. HAHN,
and ELENI MANTIS MERCADER, Administrative Patent Judges.

MANTIS MERCADER, Administrative Patent Judge.

DECISION ON APPEAL

1 The two-month time period for filing an appeal or commencing a civil action, as
recited in 37 C.F.R. § 1.304, begins to run from the decided date shown on this
page of the decision. The time period does not run from the Mail Date (paper
delivery) or Notification Date (electronic delivery).

Appeal 2009-1561
Application 10/960,775

2
STATEMENT OF THE CASE
Appellants seek our review under 35 U.S.C. § 134 of the Examiner’s final
rejection of claims 18-24. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.

INVENTION
Appellants’ claimed invention is directed to a method of speech-based
navigation for information units to predefined web sites (Spec. 2: ¶ 7). According
to the invention, a client downloads from a server a private information unit that
enables a speech input (Spec. 2: ¶ 7). A speech recognizer generates a recognition
result from an uttered speech input, and with the recognition result a link in a data
file is determined, which link is assigned to a word that correlates with the
recognition result (Spec. 2: ¶ 7).
Claim 18, reproduced below, is representative of the subject matter on
appeal:
18. A method of speech-based navigation in a communications network, the
method comprising:
downloading a web page from a web site to a client at a user location;
receiving a speech input from the user to the web page;
subdividing the speech input into representative feature vectors;
forwarding the representative feature vectors to a remote speech recognizer;
producing from the representative feature vectors a recognition result
representative of the speech input;
Appeal 2009-1561
Application 10/960,775

3
determining from the recognition result a link to an information unit in a
data file;
and
assigning the link to a word correlated with the recognition result.

THE REJECTIONS
The Examiner relies upon the following as evidence of unpatentability:
Uppaluru US 5,915,001 Jun. 22, 1999
(filed Nov. 14, 1996)

Barclay

US 5,960,399 Sep. 28, 1999
(filed Dec. 24, 1997)

The following rejections are before us for review:
1. The Examiner rejected claim 24 under 35 U.S.C. § 102(e) as being
anticipated by Uppaluru.
2. The Examiner rejected claims 18-23 under 35 U.S.C. § 103(a) as being
unpatentable over Barclay in view of Uppaluru.
Appellants argue the rejection of claims 18-23 as a group with claim 18 as
representative (Br. 4-5).2 Accordingly, claims 19-23 stand or fall with claim 18.
See 37 C.F.R. § 41.37 (c)(1)(vii) (2004).

2 Only arguments made by Appellants have been considered in this decision.
Arguments which Appellants could have made but did not make in the Brief have
not been considered and are deemed waived. See 37 C.F.R. § 41.37(c)(1)(vii)
(2004).
Appeal 2009-1561
Application 10/960,775

4
ANTICIPATION
ISSUE
Appellants contend that Uppaluru’s speaker dependent speech recognition is
not equivalent to assigning user-specific links to predefined words, as recited in
claim 24 (Br. 6). Furthermore, Appellants contend that Uppaluru teaches that in
response to a command, the system loads the requested web page and
corresponding speech training page (Br. 6). Appellants argue that this is not
equivalent to associating a link assignment with a user-identifier and transmitting
the web page and a location of a remote speech recognizer for use with the web
page, as required by claim 24 (Br. 6-7).
The Examiner responds that Uppaluru teaches in Figure 3 a personal voice
web 300 having a personal home page 301 linking to different web pages (Ans.
13). The personal voice web also contains authentication page 306 containing
authentication information including a subscriber account number (Ans. 13). Thus,
the Examiner concludes that Uppaluru teaches the limitation of “associating the
link assignment with a user identifier (IDn) at a remote data file” (Ans. 13). The
Examiner further responds that Uppaluru also teaches that as the subscriber
navigates from service page to service page in the personal voice web, the system
is able to access the corresponding speech training page using its embedded URL,
which reads on the limitation of “transmitting a web page to the local client
containing the user identification and a location of a remote speech recognizer for
use with the web page” (Ans. 13).
The issue before us, then, is as follows:
Have Appellants shown that the Examiner erred by determining that
Appeal 2009-1561
Application 10/960,775

5
Uppaluru teaches the limitations of “associating the link assignment with a user
identifier (IDn) at a remote data file” and “transmitting a web page to the local
client containing the user identification (IDn) and a location of a remote speech
recognizer for use with the web page”?

FINDINGS OF FACT
The relevant facts include the following:
1. Uppaluru teaches that a subscriber can retrieve documents using voice
commands via the Internet whereby the documents are linked using HTML
hyper-links (Abstract; col. 2, ll. 22-41 and col. 8, ll. 36-39).
2. Uppaluru teaches a personal voice web 300 for a subscriber (col. 11, ll. 12-
16) having a personal home page 301 linking to different web pages 309-327
(Fig. 3).
3. Uppaluru teaches that the personal voice web 300 contains authentication
page 306 containing authentication information including a subscriber
account number and a personal identification number that links to a voice
authentication signature MIME (col. 12, l. 66-col. 13, l. 2).
4. Uppaluru teaches that the subscriber account number along with the personal
identification number is used to retrieve the URL of the subscriber’s
personal voice web home page 301 (col. 15, l. 58-col. 16, l. 1).
5. Uppaluru teaches that once it is determined that a match exists between the
stored authentication signature at the subscriber’s personal profile page 302,
then the subscriber is given access to the personal voice web 300 (Fig. 3 and
col. 16, ll. 32-41).
Appeal 2009-1561
Application 10/960,775

6
6. Uppaluru teaches that the personal home page 301 links to web pages 309-
327 through web pages 303-305 (col. 11, ll. 25-29 and Fig. 3).
7. Uppaluru teaches that once a subscriber gains access to the personal voice
web 300 via the user identifier (IDn) (Findings of Fact 3-5 and col. 17, ll.
59-66), the subscriber navigates from service page to service page in the
personal voice web 300, and the system is able to access the corresponding
speech training page using its embedded URL (col. 17, ll. 37-45).
8. Uppaluru further teaches that the voice web browser recognizes the
command and control word utterances (subscriber’s voice commands) and
matches them against the personalized vocabulary in the corresponding
voice web speech training page for accurate speaker dependent recognition
of discrete speech (col. 18, ll. 23-29).

PRINCIPLES OF LAW
“A claim is anticipated only if each and every element as set forth in the
claim is found, either expressly or inherently described, in a single prior art
reference.” Verdegaal Bros. Inc., v. Union Oil Co. of Calif., 814 F.2d 628, 631
(Fed. Cir. 1987).

ANALYSIS
At the outset, we note that Appellants have not explained with sufficient
specificity why the limitations are not met by the Examiner’s cited portions of
Uppaluru. Nonetheless, we will address Appellants’ assertions.
Uppaluru teaches that a subscriber can retrieve documents using voice
Appeal 2009-1561
Application 10/960,775

7
commands via the Internet whereby the documents are linked using HTML hyper-
links (Finding of Fact 1). Uppaluru teaches a personal voice web 300 for a
subscriber (i.e., describing a collection of linked voice web pages—i.e., remote
data file) having a personal home page 301 linking to different web pages 309-327
(Finding of Fact 2). Thereby, Uppaluru teaches “link assignment” by associating
or assigning voice commands (i.e., predefined words) with HTML hyper-links or
web pages 309-327 (i.e., user-specific links) (Findings of Fact 1 and 2). Therefore,
contrary to Appellants’ argument (Br. 6), we find that Uppaluru discloses the
claimed assigning user-specific links to predefined words.
Uppaluru further teaches that the personal voice web 300 contains
authentication page 306 containing authentication information including a
subscriber account number and a personal identification number (i.e., IDn) that
links to a voice authentication signature MIME (Finding of Fact 3). Uppaluru
explains that the subscriber account number along with the personal identification
number is used to retrieve the URL of the subscriber’s personal voice web home
page 301 (Finding of Fact 4). Uppaluru teaches that once it is determined that a
match exists between the stored authentication signature at the subscriber’s
personal profile page 302 then the subscriber is given access to the personal voice
web 300 (Finding of Fact 5) and thereby enabling linking to web pages 309-327
(Finding of Fact 2). Uppaluru teaches that the personal home page 301 links to
web pages 309-327 through web pages 303-305 (Finding of Fact 6). Thus, we
again disagree with Appellants’ assertion (Br. 6) against Uppaluru’s teaching
associating a link assignment with a user-identifier. Based on the findings above,
we agree with the Examiner’s conclusion (Ans. 13) that Uppaluru teaches
Appeal 2009-1561
Application 10/960,775

8
associating the link assignment (Findings of Fact 1 and 2) with a user identifier
(IDn) at a remote data file (Findings of Fact 3-6).
Uppaluru teaches that once a subscriber gains access to the personal voice
web 300 via the user identifier (IDn), the subscriber navigates from service page to
service page (i.e., in Fig. 3 web pages 302-327) in the personal voice web 300, and
the system is able to access the corresponding speech training page (i.e., in Fig. 4
speech training pages 402-427) using its embedded URL (i.e., the transmitted web
page includes the URL or the location of the remote speech training page) (Finding
of Fact 7). As stated by the Examiner (Ans. 13), Uppaluru further teaches that the
voice web browser recognizes the command and control word utterances
(subscriber’s voice commands) and matches them against the personalized
vocabulary in the corresponding voice web speech training page for accurate
speaker dependent recognition of discrete speech (i.e., the remote speech training
page is used with its respective web page for accurate command recognition)
(Finding of Fact 8).
Thus, contrary to Appellants’ arguments (Br. 6-7), Uppaluru teaches
transmitting a web page (i.e., in Fig. 3 web page 322) to the local client (i.e.,
personal home page 301) containing the user identification (IDn) (i.e., via the
authentication process through web pages 302 and 306 recognizing (IDn) and
allowing personal home page 301 to access web page 322) and a location of a
remote speech recognizer for use with the web page (i.e., URL location of web
page 422) (Findings of Fact 6-8).
For the foregoing reasons, Appellants have not persuaded us that the
Examiner erred in rejecting claim 24. Accordingly, we sustain the Examiner’s
Appeal 2009-1561
Application 10/960,775

9
rejection of claim 24.
CONCLUSION
Appellants have not shown that the Examiner erred by determining that
Uppaluru teaches the limitations of “associating the link assignment with a user
identifier (IDn) at a remote data file” and “transmitting a web page to the local
client containing the user identification (IDn) and a location of a remote speech
recognizer for use with the web page.” Thus, the Examiner appropriately rejected
claim 24 under 35 U.S.C. § 102(e).

OBVIOUSNESS
ISSUE
Appellants contend that Barclay does not teach or suggest “determining a
link to an information unit in a data file” and then “assigning the link to a word
correlated with the recognition result” as required by claim 18 (Br. 5).
The Examiner responds that Barclay’s Figure 4, indicates that the "Decoder"
(i.e., "speech recognizer") produces a recognition result representative of the
speech input which is used to determine a link to an information unit (i.e., Web
page) in a data file (Netscape) (Ans. 8-9). The Examiner states that Barclay’s
statement of “the applet controls the plug-in through various interfaces function,
for example the applet may command the plug-in to being the speech processing as
well as passing keyword-value pairs for control information to accompany the
speech processing” (col. 8, ll. 60-64) reads on the link being assigned to a word
correlated with the recognition result as recited in claim 18 (Ans. 9).
The issue before us, then, is as follows:
Appeal 2009-1561
Application 10/960,775

10
Have Appellants shown that the Examiner erred by determining that Barclay
teaches “determining from the recognition result a link to an information unit in a
data file” and “assigning the link to a word correlated with the recognition result”
as recited in claim 18?

FINDINGS OF FACT
The relevant facts include the following:
1. The Examiner determined (Ans. 8) that Barclay’s Figure 4, indicates that the
“Decoder” (i.e., “speech recognizer”) produces a recognition result
representative of the speech input which is used to determine a link to an
information unit (i.e., Web page) in a data file (Netscape) (col. 8, ll. 48-64).
2. The Examiner further determined (Ans. 9) that Barclay’s statement of “the
applet controls the plug-in through various interfaces function, for example
the applet may command the plug-in to begin the speech processing as well
as passing keyword-value pairs for control information to accompany the
speech processing” (col. 8, ll. 60-64) reads on the link being assigned to a
word correlated with the recognition result, as recited in claim 18.
3. Claim 18 recites the term “link”—rather than the term hyperlink.
4. The term “link” is defined, in pertinent part, as “connection.” Encantra
Online Dictionary, at
http://encarta.msn.com/encnet/features/dictionary/DictionaryResults.aspx?re
fid=1861698851(last visited March 19, 2009).
5. Barclay teaches that when the user utters the words “I want to fly from
Boston to Denver tomorrow,” the server would recognize and interpret the
Appeal 2009-1561
Application 10/960,775

11
speech and return the result for example of “ORIGIN: Boston” (col. 9, ll. 5-
25).
PRINCIPLES OF LAW
The Examiner bears the initial burden of presenting a prima facie case of
obviousness. In re Oetiker, 977 F.2d 1443, 1445 (Fed. Cir. 1992). If that burden is
met, then the burden shifts to the Appellants to overcome the prima facie case with
argument and/or evidence. Id.
During ex parte prosecution, claims must be interpreted as broadly as their
terms reasonably allow since Applicants have the power during the administrative
process to amend the claims to avoid the prior art. In re Zletz, 893 F.2d 319, 322
(Fed. Cir. 1989).
Although claims are interpreted in light of the specification, limitations from
the specification are not read into the claims. In re Van Geuns, 988 F.2d 1181,
1184 (Fed. Cir. 1993).

ANALYSIS
As stated supra, the Examiner determined that Barclay’s Figure 4, indicates
that the “Decoder” (i.e., “speech recognizer”) produces a recognition result
representative of the speech input which is used to determine a link to an
information unit (i.e., Web page) in a data file (Netscape) (Finding of Fact 1). The
Examiner further determined that Barclay’s statement of “the applet controls the
plug-in through various interfaces function, for example the applet may command
the plug-in to begin the speech processing as well as passing keyword-value pairs
for control information to accompany the speech processing” reads on the link
Appeal 2009-1561
Application 10/960,775

12
being assigned to a word correlated with the recognition result as recited in claim
18 (Finding of Fact 2) (emphasis added).
We agree with the Examiner’s findings of facts and conclusions as set out in
the Answer and adopt them as our own. We add the following primarily for
emphasis.
At the outset, we note that claim 18 recites the term “link”—rather than the
term hyperlink (Finding of Fact 3). The term “link” is defined, in pertinent part, as
“connection” (Finding of Fact 4).
Barclay teaches that when the user utters the words “I want to fly from
Boston to Denver tomorrow,” the server would recognize and interpret the speech
and return the result for example of “ORIGIN: Boston” (i.e., Examiner’s reference
to keyword-value pair) (Finding of Fact 5). The Examiner interpreted the word
“link” according to its customary meaning of “connection” wherein there is a
determination from the recognition result (i.e., the recognition of the word
“Boston” as “Origin”—the result is “Origin”), a link (i.e., connection) to an
information unit (i.e., Web Page), and assigning the link (i.e., the portion of the
Web page reciting “Origin”) to a word (i.e., Boston) correlated with the
recognition result (“Origin”).
As stated supra, during ex parte prosecution, claims must be interpreted as
broadly as their terms reasonably allow (i.e., link interpreted as connected under
the broadest reasonable interpretation) since Applicants have the power during the
administrative process to amend the claims to avoid the prior art. Zletz, 893 F.2d
at 322. Furthermore, although claims are interpreted in light of the specification, a
limitation from the specification (i.e., hyperlink) is not read into the claims. Van
Appeal 2009-1561
Application 10/960,775

13
Geuns, 988 F.2d at 1184. Here, based on the broadest reasonable interpretation of
the term “link,” and contrary to Appellants’ assertion (Br. 6), Barclay does teach
“determining a link” and then “assigning the link to a word” as required by claim
18.
For the foregoing reasons, Appellants have not persuaded us that the
Examiner erred in rejecting claim 18 and claims 19-23 which fall with claim 18.
Accordingly, we sustain the Examiner’s rejection of claims 18-23.

CONCLUSION
Appellants have not shown that the Examiner erred by determining that
Barclay teaches “determining from the recognition result a link to an information
unit in a data file” and “assigning the link to a word correlated with the recognition
result” as recited in claim 18. Thus, the Examiner appropriately rejected claims
18-23 under 35 U.S.C. § 103(a).
ORDER
The decision of the Examiner to reject claim 24 under 35 U.S.C.
§ 102(e) is affirmed. The decision of the Examiner to reject claims 18-23
under 35 U.S.C. § 103(a) is also 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
Appeal 2009-1561
Application 10/960,775

14
tdl

BROMBERG & SUNSTEIN LLP
125 SUMMER STREET
BOSTON, MA 02110-1618