GREGORY CIURPITA
Pilots at Bartle Rd, East Millstone, NJ

2003002, Jan 30, 2003

May 2, 2001

09/846200

Gregory Ciurpita - Somerset NJ, US
Sunil Gupta - Edison NJ, US
Prabhu Ragavan - Edison NJ, US

G10L015/00, G10L017/00

704/246000

A method and system of recognizing speech based in part on an observation that a speaker naturally pauses and speaks smaller subgroups of speech units or digits that form part of a complete longer speech sequence. In the method, subgroups of speech units are processed by the system between a human's natural pauses. This pause is detected by the system and the subgroup is processed in order to provide a recognition result, which is a best representation of the input subgroup. The recognition result is immediately repeated back to the user for verification. The user is prompted to repeat a subgroup for re-recognition and re-verification if a rejection criteria is met; otherwise the processing steps are repeated for remaining subgroups until it has been determined that the complete speech sequence has been accurately recognized.

5832077, Nov 3, 1998

Jan 16, 1997

8/783812

Gregory Ciurpita - Somerset NJ

Lucent Technologies Inc. - Murray Hill NJ

H04M 356

379388

A microphone/loudspeaker includes a microphone arrangement and a loudspeaker such that the microphone/loudspeaker operates in a quiescent mode and shifts to a speakerphone mode in response to detecting a battery voltage on a telephone line. In the quiescent mode the microphone arrangement and the loudspeaker operate substantially independent of each other while in the speakerphone mode they interact on the basis of their respective signal strengths. In a system, the microphone/loudspeaker is connected to other microphone/loudspeakers so that the microphone arrangement of one microphone/loudspeaker couples in tandem to the microphone arrangement of other microphone/loudspeakers, and the loudspeaker of one microphone/loudspeaker couples in tandem to the loudspeaker of other microphone/loudspeakers.

5625697, Apr 29, 1997

May 8, 1995

8/436671

Donald J. Bowen - Madison NJ
Gregory Ciurpita - Somerset NJ

Lucent Technologies Inc. - Murray Hill NJ

H04R 300

381 92

A microphone selection process is made relatively constant in terms of processing requirements through use of combination values which provide a measure of the quality of the speech signal received in each one of a plurality of microphones. Such combination values are derived in a manner which provides an indication of which one of the plurality of microphones, that provide area coverage for a conference room, for example, best receives the speech signal. In operation, the microphone selection process selects the microphone which best receives the speech signal by comparing a signal energy value received in each one of the microphones to that received in each of the other microphones. More specifically, pairs of microphones are examined, to determine the direction of origin for the speech, by seeking a microphone pair where the speech is strong in the microphone directed toward the source of speech, and weak in an associated microphone directed away from the source of speech. Since a null, which is located at the back of each microphone, is narrower than a main beam or sensitivity pattern, which is located at the front of each microphone, this null is more sensitive to and therefore a better indicator of the direction of origin for the speech than is the main beam.