Ex Parte Tanner et al

Board of Patent Appeals and Interferences

May 9, 2012

10355148 (B.P.A.I. May. 9, 2012)

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.
10/355,148 01/31/2003 Marc D. Tanner 0005836USU/4247 3795
7590 05/10/2012
Charles N.J. Ruggiero
Ohlandt, Greeley, Ruggiero & Perle, L.L.P.
10th Floor
One Landmark Square
Stamford, CT 06901-2682
EXAMINER
SAINT CYR, LEONARD
ART UNIT PAPER NUMBER
2626
MAIL DATE DELIVERY MODE
05/10/2012 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 MARC D. TANNER and ERIN M. PANTTAJA

________________

Appeal 2009-014380
Application 10/355,148
Technology Center 2600
________________

Before MAHSHID D. SAADAT, DEBRA K. STEPHENS, and
JASON V. MORGAN, Administrative Patent Judges.

MORGAN, Administrative Patent Judge.

DECISION ON APPEAL
Appeal 2009-014380
Application 10/355,148

2
STATEMENT OF THE CASE
Introduction
This is an appeal under 35 U.S.C. § 134(a) from the Examiner’s final
rejection of claims 1 – 73. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm-in-part.
Invention
The invention relates to the automated recognition of proper nouns
using pronunciation specific to one or more geographical areas (Spec.
¶ [0001]). This includes augmenting the pronunciation proper nouns or
names in the natural language of the speech recognition system with at least
one “native” pronunciation in another natural language (Abstract). To
maximize recognition, preferably the pronunciations are predicted based on
information not available to the speech recognition system, such as a
location derived from a telephone number or postal address (id.).
Exemplary Claims
1. A method of improving automated recognition of proper
nouns spoken by users, comprising:
predicting at least one native pronunciation, in at least one
natural language, of a proper noun stored as text, based on
information associated with the proper noun but not derived
from the text of the proper noun; and
augmenting a speech recognizer, that is not configured to
recognize words in the at least one natural language, using the
at least one native pronunciation of the proper noun.
(Emphasis added).
59. A system to automatically recognize proper nouns spoken
by users, comprising:
Appeal 2009-014380
Application 10/355,148

3
a speech recognizer configured to recognize words in at least a
first natural language, but not a second natural language; and
a processor, coupled to said speech recognizer, to predict at
least one native pronunciation, in at least the second natural
language, of a proper noun and to augment the speech
recognizer to recognize the at least one native pronunciation of
the proper noun.
Evidence and Rejections
The Examiner rejects claims 1 – 4, 9 – 13, 18 – 24, 27 – 33, 38, 39,
44, 45, 48 – 51, 54 – 66, and 70 – 73 under 35 U.S.C. § 102(b) as being
anticipated by Kiraz (US 6,272,464 B1) (Ans. 3 – 17).1
ISSUES
1. Did the Examiner err in finding that Kiraz discloses “predicting at
least one native pronunciation, in at least one natural language, of a proper
noun stored as text, based on information associated with the proper noun
but not derived from the text of the proper noun,” as recited in claim 1?
2. Did the Examiner err in finding that Kiraz discloses “a speech
recognizer configured to recognize words in at least a first natural language,
but not a second natural language” and a processor “to predict at least one
native pronunciation, in at least the second natural language, of a proper
noun and to augment the speech recognizer to recognize the at least one
native pronunciation of the proper noun,” as recited in claim 59?

1 The Examiner also rejects dependent claims 5 – 8, 14 – 17, 25, 26, 34 – 37,
40 – 43, 46, 47, 52, 53, and 67 – 69 under 35 U.S.C. § 103(a) as being
unpatentable over Kiraz in combination with other art (Ans. 18 – 27).
Appellants’ submissions with respect to these other claims are either (1)
moot in light of our findings and conclusions below or (2) lack specificity
and fail to present separate arguments for patentability (see App. Br. 17).
Appeal 2009-014380
Application 10/355,148

4
ANALYSIS
Claim 1
The Examiner finds that because Kiraz discloses generating
pronunciations of a given name based on a name language origin identifier,
Kiraz thus discloses predicting at least one native pronunciation of a proper
noun based on “based on information associated with the proper noun but
not derived from the text of the proper noun” (see Ans. 3 and 27 – 28; see
also Kiraz col. 3, ll. 43 – 45; col. 5, ll. 34 – 43). Appellants contend that the
Examiner erred, pointing to evidence that Kiraz obtains the name language
origin identifier based on “a statistical analysis of successive letter
sequences which are contained in the given name” (see App. Br. 9 (citing
Kiraz col. 4, ll. 54 – 55); see also Kiraz Fig. 2).
We agree with Appellants that this evidence shows that the name
language origin identifier in Kiraz is information derived from the text of the
proper noun whose pronunciation is to be predicted. The letter sequences are
text of the proper noun. Thus, performing statistical analysis on the letter
sequences to provide a name language origin identifier is a method of
deriving information from the text of the proper noun.
The Examiner has not shown that Kiraz discloses any methods of
obtaining a name language origin identifier without deriving this information
from the text of the proper noun, or that any of the other references cure this
deficiency. Accordingly, we agree with Appellants that the Examiner erred
in rejecting claim 1, as well as claims 2 – 58, and 70 – 72, which contain the
same or similar recitations.
Appeal 2009-014380
Application 10/355,148

5
Claim 59
We agree with the Examiner’s finding that Kiraz discloses both the
speech recognizer and the processor of claim 59 (Ans. 14 – 15 and 28 – 29).
Kiraz first discloses a “method or apparatus for use in performing speech
recognition of speech utterances which include the proper name by
individuals within a given population of speakers” (Kiraz col. 3, ll. 59 – 62).
This describes a speech recognizer configured to recognize words in at least
a first natural language (i.e., a language of the given population of speakers),
but not a second natural language (i.e., a language other than a language of
the given population of speakers).
Kiraz also discloses “identifying one or more of a plurality of
languages as a potential origin of the proper name” and “generating a
plurality of plausible pronunciations of the given name . . . based on the one
or more identified languages” (Kiraz col. 3, l. 63 – col 4, l. 1). This describes
a processor to predict a native pronunciation in the second natural
language—the identified potential origin being a language other than a
language of the given population of speakers—and to augment the speech
recognizer using the predicted pronunciation.
Kiraz gives an example of the prediction process with the Chinese
name “Quan,” which is commonly mispronounced [kwan] (Kiraz col. 2, ll.
44 – 47). The speech recognizer in Kiraz recognizes the mispronunciation,
which is based on the language of the given population of speakers. Kiraz
discloses also identifying that the name is Chinese in origin, and augmenting
the speech recognizer to recognize the native Mandarin and Cantonese
pronunciations [ɕüen] and [ɕʉn] (Kiraz col. 2, ll. 44 – 52).
Appeal 2009-014380
Application 10/355,148

6
Appellants argue that Kiraz does not disclose “a speech recognizer
that predicts pronunciations of a word in a language in which it is not
configured to recognize words” (App. Br. 14). However, the claim recites
both a speech recognizer and a processor. The processor, not the speech
recognizer, performs the prediction. Thus, this argument is not
commensurate with the scope of the claimed invention, and not persuasive
of error.
Appellants’ other arguments merely point to what claim 59 recites
without showing how the recitations distinguish the claimed features over
Kiraz (App. Br. 13 – 14). Accordingly, we do not find error in the
Examiner’s rejection of claim 59, as well as dependent claims 60 – 69 and
73, which are not argued separately or with sufficient specificity.
DECISION
We reverse the Examiner’s decision rejecting claims 1 – 4, 9 – 13, 18
– 24, 27 – 33, 38, 39, 44, 45, 48 – 51, 54 – 58, and 70 – 72 under 35 U.S.C.
§ 102(b) and claims 5 – 8, 14 – 17, 25, 26, 34 – 37, 40 – 43, 46, 47, 52, and
53 under 35 U.S.C. § 103(a).
We affirm the Examiner’s decision rejecting claims 59 – 66 and 73
under 35 U.S.C. § 102(b) and claims 67 – 69 under 35 U.S.C. § 103(a).
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-IN-PART

msc