Ronald M Cole

US Patent:

20060080301, Apr 13, 2006

Filed:

Sep 15, 2005

Appl. No.:

11/227878

Inventors:

Ronald Cole - Arlington VA, US

International Classification:

G06F 17/30

US Classification:

707003000

Abstract:

Computer users frequently need to search for items, such as inventory, based on specific criteria. The specific criteria used are often dependent upon the task they are performing. For example, if they are looking for books they may search by ‘title’. If they are looking for cars they may look by ‘engine type’. This invention allows a user to create the search screens (with places to enter the values for the pertinent fields) automatically by the use of a software wizard. This invention extracts information about the database, and then creates the steps necessary for the person creating the search screen to create it. The person creating the search screen can control what fields the search screen user will be able to search on, how the data will be filtered and/or manipulated, how it will be formatted, and how it will be presented. This provides a tremendous productivity boost to the search screen user, since much of the filtering and decision-making will have been done before hand. Also, by using a wizard to allow the rapid creation of searches, the door is opened for ‘rapid application development.’ This allows non-programmers the ability to create, rapidly deploy, test and refine their own searches.

US Patent:

56218575, Apr 15, 1997

Filed:

Dec 20, 1991

Appl. No.:

7/811819

Inventors:

Ronald A. Cole - Portland OR
Mark A. Fanty - Portland OR

Assignee:

Oregon Graduate Institute of Science and Technology - Portland OR

International Classification:

G10L 506
G10L 900

US Classification:

395 241

Abstract:

Improved system and method for speaker-independent speech token recognition are described. The system is neural network-based and involves processing a sequence of spoken utterances, e. g. separately articulated letters of a name, to identify the same based upon a highest probability match of each utterance with learned speech tokens, e. g. the letters of the English language alphabet, and based upon a highest probability match of the uttered sequence with a defined utterance library, e. g. a list of names. First, the spoken utterance is digitized or captured and processed into a spectral representation. Second, discrete time frames of the DFT are classified phonetically. Third, the time-frame outputs are used by a modified Viterbi search to locate segment boundaries, near which such segment boundaries lies the information that is needed to discriminate letters. Fourth, the segmented or bounded representation is reclassified using such information into individual hypothesized letters. Fifth, successive, hypothesized letter scores are analyzed to obtain a high probability match with a spelled word within the utterance library.