LONG NGUYEN
Broker in Medford, MA

6618726, Sep 9, 2003

Aug 28, 2001

09/942276

Sean C. Colbath - Boston MA
David Stallard - Cambridge MA
Cecile Pham - Norwood MA
Long Nguyen - Medford MA

Genuity Inc. - Burlington MA
Verizon Corporate Services Group Inc. - New York NY

G06F 1750

707 6, 707 10, 704231

A voice activated and operated Internet web browser where voiced utterances are digitized and the digitized representation is input to a speech recognition program is disclosed. The speech recognition program determines the most likely set of words from a stored vocabulary making up the utterances. The words are analyzed and searched against a database of topics or micro-domain areas. If a match occurs the words are analyzed by a subprogram associated with the specific topic area and key words are identified and other words discarded. The key words are input to a search engine or the equivalent. The search engine returns information associated with the topic. If no topic match is found a general search on the recognized words is performed. The recognized words are fed to a identification module where key words are extracted that are associated with names of people, organizations, locations, and companies; times; money amounts; percentages; dates; and not-a-name. The extraneous not-a-name words are discarded and the remaining key words are passed to a search engine.

6311182, Oct 30, 2001

Nov 17, 1997

8/971606

Sean C. Colbath - Boston MA
David Stallard - Cambridge MA
Cecile Pham - Norwood MA
Long Nguyen - Medford MA

Genuity Inc. - Burlington MA
GTE Service Corporation - Irving TX

G06F 1730

707 6

A voice activated and operated Internet web browser where voiced utterances are digitized and the digitized representation is input to a speech recognition program is disclosed. The speech recognition program determines the most likely set of words from a stored vocabulary making up the utterances. The words are analyzed and searched against a database of topics or micro-domain areas. If a match occurs the words are analyzed by a subprogram associated with the specific topic area and key words are identified and other words discarded. The key words are input to a search engine or the equivalent. The search engine returns information associated with the topic. If no topic match is found a general search on the recognized words is performed. The recognized words are fed to a identification module where key words are extracted that are associated with names of people, organizations, locations, and companies; times; money amounts; percentages; dates; and not-a-name. The extraneous not-a-name words are discarded and the remaining key words are passed to a search engine.

5621859, Apr 15, 1997

Jan 19, 1994

8/183719

Richard M. Schwartz - Sudbury MA
Long Nguyen - Medford MA

BBN Corporation - Cambridge MA

G10L 506, G10L 900

395 265

The invention provides a method of large vocabulary speech recognition that employs a single tree-structured phonetic hidden Markov model (HMM) at each frame of a time-synchronous process. A grammar probability is utilized upon recognition of each phoneme of a word, before recognition of the entire word is complete. Thus, grammar probabilities are exploited as early as possible during recognition of a word. At each frame of the recognition process, a grammar probability is determined for the transition from the most likely preceding grammar state to a set of words that share at least one common phoneme. The grammar probability is combined with accumulating phonetic evidence to provide a measure of the likelihood that a state in the HMM will lead to the word most likely to have been spoken. In a preferred embodiment, phonetic context information is exploited, even before the complete context of a phoneme is known. Instead of an exact triphone model, wherein the phonemes previous and subsequent to a phoneme are considered, a composite triphone model is used that exploits partial phonetic context information to provide a phonetic model that is more accurate than aphonetic model that ignores context.