Lan Zhang

7890328, Feb 15, 2011

Sep 7, 2006

11/470685

Harry Blanchard - Rumson NJ, US
Steven Lewis - Middletown NJ, US
Shankarnarayan Sivaprasad - Highland Park NJ, US
Lan Zhang - Freehold NJ, US

AT&T Intellectual Property II, L.P. - New York NY

G10L 15/18
G10L 15/06
G10L 21/00

704257, 704244, 704270

Disclosed herein are methods and systems for recognizing speech. A method embodiment comprises comparing received speech with a precompiled grammar based on a database and if the received speech matches data in the precompiled grammar then returning a result based on the matched data. If the received speech does not match data in the precompiled grammar, then dynamically compiling a new grammar based only on new data added to the database after the compiling of the precompiled grammar. The database may comprise a directory of names.

8023627, Sep 20, 2011

Dec 27, 2006

11/616728

Harry Blanchard - Rumson NJ, US
Steven H. Lewis - Middletown NJ, US
Gregory Pulz - Cranbury NJ, US
Lan Zhang - Freehold NJ, US

AT&T Intellectual Property II, L.P. - Atlanta GA

H04M 11/00

379 9301, 379 9001, 34087007

A method and apparatus for retrieving information from household appliances are disclosed. For example, the method receives a request from a user of an appliance. The method then receives stored information from the appliance as one or more Dual Tone Multi-Frequency (DTMF) signals.

8150020, Apr 3, 2012

Apr 4, 2007

11/732742

Harry E. Blanchard - Rumson NJ, US
Steven H. Lewis - Middletown NJ, US
Gregory Pulz - Cranbury NJ, US
Lan Zhang - Freehold NJ, US

AT&T Intellectual Property II, L.P. - Atlanta GA

H04M 5/00

37926503, 37926501

In a method for use in Interactive Voice Response (IVR) systems, an IVR is configured such that voice prompts provided to a caller vary systematically and caller hang-up activity is attributed to the particular version of a prompt played to the caller. Voice prompt modifications are chosen based on how long the caller willingly listens to the voice prompt. A relatively short time listening to a prompt before hang-up, indicates dissatisfaction with that prompt. The system compares caller hang-up rates for each of the phrases in a prompt and chooses the optimum solution, which is the variant of the prompt with the longest caller listening time. Optionally, the system compares hang up rates to a threshold and contingent on that comparison chooses an alternative version of the prompt.

8204205, Jun 19, 2012

Feb 8, 2007

11/704000

Harry Blanchard - Rumson NJ, US
Steven Lewis - Middletown NJ, US
Gregory Pulz - Cranbury NJ, US
Lan Zhang - Wayne PA, US

AT&T Intellectual Property II, L.P. - Atlanta GA

H04M 5/00

37926501, 37926502, 37926507, 37926601, 37926602, 37926603, 704270, 704251, 704257, 704275

The problem of an automated telephony system having only a single way to measure a caller's difficulty in processing a transaction when using the system is solved by selecting from at least two thresholds for measuring the caller's difficulty. Having a choice of alternative thresholds to measure the caller's difficulty, referred to as inefficiency thresholds, allows for a more responsive approach to managing an automated telephony system. This choice of inefficiency thresholds overcomes multiple problems of systems only using one inefficiency threshold, such as: (1) transferring a call to a live agent to quickly, (2) transferring a call to a live agent to slowly, and (3) giving a caller a feeling that the automated telephony system is unresponsive to his or her needs. Having multiple inefficiency thresholds from which to choose allows for a more dynamic response to a diverse audience of callers and to different needs of a single caller during the various parts of a call.

8229746, Jul 24, 2012

Feb 14, 2011

13/026884

Harry Blanchard - Rumson NJ, US
Steven Lewis - Middletown NJ, US
Shankarnarayan Sivaprasad - Highland Park NJ, US
Lan Zhang - Freehold NJ, US

AT&T Intellectual Property II, L.P. - Atlanta GA

G10L 15/18
G10L 15/06
G10L 21/00

704257, 704244, 704270

Disclosed herein are methods and systems for recognizing speech. A method embodiment comprises comparing received speech with a precompiled grammar based on a database and if the received speech matches data in the precompiled grammar then returning a result based on the matched data. If the received speech does not match data in the precompiled grammar, then dynamically compiling a new grammar based only on new data added to the database after the compiling of the precompiled grammar. The database may comprise a directory of names.

8478593, Jul 2, 2013

Jul 18, 2012

13/552218

Harry Blanchard - Rumson NJ, US
Steven Lewis - Middletown NJ, US
Shankarnarayan Sivaprasad - Highland Park NJ, US
Lan Zhang - Freehold NJ, US

AT&T Intellectual Property II, L.P. - Atlanta GA

G10L 15/18
G10L 15/06
G10L 21/00

704257, 704244, 704270

Disclosed herein are methods and systems for recognizing speech. A method embodiment comprises comparing received speech with a precompiled grammar based on a database and if the received speech matches data in the precompiled grammar then returning a result based on the matched data. If the received speech does not match data in the precompiled grammar, then dynamically compiling a new grammar based only on new data added to the database after the compiling of the precompiled grammar. The database may comprise a directory of names.

8488771, Jul 16, 2013

Feb 23, 2012

13/402958

Harry E. Blanchard - Rumson NJ, US
Steven H. Lewis - Middletown NJ, US
Gregory Pulz - Cranbury NJ, US
Lan Zhang - Freehold NJ, US

AT&T Intellectual Property II, L.P. - Atlanta GA

H04M 3/00

37926503

In a method for use in Interactive Voice Response (IVR) systems, an IVR is configured such that voice prompts provided to a caller vary systematically and caller hang-up activity is attributed to the particular version of a prompt played to the caller. Voice prompt modifications are chosen based on how long the caller willingly listens to the voice prompt. A relatively short time listening to a prompt before hang-up, indicates dissatisfaction with that prompt. The system compares caller hang-up rates for each of the phrases in a prompt and chooses the optimum solution, which is the variant of the prompt with the longest caller listening time. Optionally, the system compares hang up rates to a threshold and contingent on that comparison chooses an alternative version of the prompt.

8532279, Sep 10, 2013

May 16, 2012

13/472908

Harry Blanchard - Rumson NJ, US
Steven Lewis - Middletown NJ, US
Gregory Pulz - Cranbury NJ, US
Lan Zhang - Wayne PA, US

AT&T Intellectual Property II, L.P. - Atlanta GA

H04M 5/00

37926501, 37926507, 37926601, 37926602, 37926603

The problem of an automated telephony system having only a single way to measure a caller's difficulty in processing a transaction when using the system is solved by selecting from multiple thresholds for measuring the caller's difficulty. Having a choice of alternative thresholds to measure the caller's difficulty, referred to as inefficiency thresholds, allows for a more responsive approach to managing an automated telephony system. This choice of inefficiency thresholds overcomes multiple problems of systems only using one inefficiency threshold, such as: (1) transferring a call to a live agent too quickly, (2) transferring a call to a live agent too slowly, and (3) giving a caller a feeling that the automated telephony system is unresponsive to his or her needs. Having multiple inefficiency thresholds allows for a more dynamic response to a diverse audience of callers and serves different needs of a caller during various parts of a call.