Robert S Roth

US Patent:

6912498, Jun 28, 2005

Filed:

May 2, 2001

Appl. No.:

09/845769

Inventors:

Daniell Stevens - Somerville MA, US
Robert Roth - Newtonville MA, US
Joel M. Gould - Winchester MA, US
Michael J. Newman - Somerville MA, US
Dean Sturtevant - Waltham MA, US
Charles E. Ingold - Bedford MA, US
David Abrahams - Cambridge MA, US
Allan Gold - Acton MA, US

Assignee:

ScanSoft, Inc. - Peabody MA

International Classification:

G10L015/04

US Classification:

704235, 704251, 704256, 704244

Abstract:

Correcting incorrect text associated with recognition errors in computer-implemented speech recognition includes receiving a selection of a word from a recognized utterance. The selection indicates a bound of a portion of the recognized utterance to be corrected. A first recognition correction is produced based on a comparison between a first alternative transcript and the recognized utterance. A second recognition correction is produced based on a comparison between a second alternative transcript and the recognized utterance. The duration of the first recognition correction differs from the duration of the second recognition correction. A portion of the recognition result that is replaced with one of the first recognition correction and the second recognition correction. includes at one bound a word indicated by the selection and extends for the duration of the one of the first recognition correction and the second recognition correction with which the portion is replaced.

US Patent:

7003456, Feb 21, 2006

Filed:

Jun 12, 2001

Appl. No.:

09/878173

Inventors:

Laurence S. Gillick - Newton MA, US
Robert Roth - Newtonville MA, US
Linda Manganaro - Arlington MA, US
Barbara R. Peskin - Berkeley CA, US
David C. Petty - Cambridge MA, US
Ashwin Rao - Redmond WA, US

International Classification:

G10L 15/26

US Classification:

704235, 704246, 704239, 704256, 704252

Abstract:

A computer-based method of routing a message to a system includes receiving a message, and processing the message using large-vocabulary continuous speech recognition to generate a string of text corresponding to the message. The method includes generating a confidence estimate of the string of text corresponding to the message and comparing the confidence estimate to a predetermined threshold. If the confidence estimate satisfies the predetermined threshold, the string of text is forwarded to the system. If the confidence estimate does not satisfy the predetermined threshold, the information relating to the message is forwarded to a transcriptionist. The message may include one or more utterances. Each utterance in the message may be separately or jointly processed. In this way, a confidence estimate may be generated and evaluated for each utterance or for the whole message.

US Patent:

7315818, Jan 1, 2008

Filed:

May 11, 2005

Appl. No.:

11/126271

Inventors:

Daniell Stevens - Somerville MA, US
Robert Roth - Newtonville MA, US
Joel M. Gould - Winchester MA, US
Michael J. Newman - Somerville MA, US
Dean Sturtevant - Waltham MA, US
Charles E. Ingold - Bedford MA, US
David Abrahams - Cambridge MA, US
Allan Gold - Acton MA, US

Assignee:

Nuance Communications, Inc. - Burlington MA

International Classification:

G10L 15/26

US Classification:

704235, 704260, 704258, 704256, 704251

Abstract:

New techniques and systems may be implemented to improve error correction in speech recognition. These new techniques and systems may be implemented to correct errors in speech recognition systems may be used in a standard desktop environment, in a mobile environment, or in any other type of environment that can receive and/or present recognized speech.

US Patent:

7676367, Mar 9, 2010

Filed:

Feb 20, 2004

Appl. No.:

10/783518

Inventors:

Robert Roth - Newton MA, US
Arkady Khasin - Acton MA, US
Laurence S. Gillick - Newton MA, US

Assignee:

Voice Signal Technologies, Inc. - Woburn MA

International Classification:

G10L 15/04
G10L 15/00

US Classification:

704252, 704251, 704231

Abstract:

A method of constructing a list of alternate transcripts from a recognized transcript includes generating a list of close call records, matching partial sub-histories from the recognized transcript with one of the history pairs stored in each of the records, and substituting the other of the history pairs for the partial sub-history of the recognized transcript. A close call record is generated each time a pair of partial hypotheses attempt to seed a common word. Each close call record includes history information and scoring information associated with a particular pair of partial hypotheses seeding a common word. Alternate transcripts are constructed by substituting close call histories for partial histories of the recognized transcripts, and also by substituting close call histories for partial histories of other alternate transcript.

US Patent:

20120327075, Dec 27, 2012

Filed:

Nov 17, 2010

Appl. No.:

13/514713

Inventors:

Lyubomir Georgiev Zagorchev - Burlington MA, US
Elizabeth Anne Morre - 's-Hertogenbosch, NL
Matthew A. Garlinghouse - Hanover NH, US
Robert M. Roth - Hanover NH, US
Thomas W. McAllister - Wilmot NH, US
Reinhard Kneser - Aachen, DE
Dieter Geller - Aachen, DE
Jochen Peters - Aachen, DE
Juergen Weese - Aachen, DE
Yuechen Qian - Briarcliff Manor NY, US

Assignee:

TRUSTEES OF DARTMOUTH COLLEGE - HANOVER NH
KONINKLIJKE PHILIPS ELECTRONICS N.V. - EINDHOVEN

International Classification:

G06T 7/00

US Classification:

345419

Abstract:

A system and method for automatic segmentation, performed by selecting a deformable model of an anatomical structure of interest imaged in a volumetric image, the deformable model formed of a plurality of polygons including vertices and edges, displaying the deformable model on a display, detecting a feature point of the anatomical structure of interest corresponding to each of the plurality of polygons and adapting the deformable model by moving each of the vertices toward the corresponding feature points until the deformable model morphs to a boundary of the anatomical structure of interest, forming a segmentation of the anatomical structure of interest.

US Patent:

20130066189, Mar 14, 2013

Filed:

Feb 2, 2011

Appl. No.:

13/580390

Inventors:

Lyubomir Georgiev Zagorchev - Burlington MA, US
Reinhard Kneser - Aachen, DE
Dieter Geller - Aachen, DE
Yuechen Qian - Briarcliff Manor NY, US
Juergen Weese - Aachen, DE
Matthew A. Garlinghouse - Hanover NH, US
Robert M. Roth - Lebanon NH, US
Thomas W. McAllister - Wilmot NH, US

Assignee:

KONINKLIJKE PHILIPS ELECTRONICS N.V. - Eindhoven

International Classification:

A61B 6/00

US Classification:

600407

Abstract:

A system and method for identifying an abnormality of an anatomical structure. The system and method segments, using a processor, the anatomical structure imaged in a volumetric image of a plurality of control patients to produce a control segmentation of the anatomical structures of each of the control patients, obtains a normative dataset by extracting a statistical representation of a morphology of the control segmentations, segments the anatomical structure of a patient being analyzed for abnormalities to produce a patient segmentation and compares the patient segmentation to the normative dataset obtained from the control segmentations.

US Patent:

20130289994, Oct 31, 2013

Filed:

Apr 26, 2012

Appl. No.:

13/456959

Inventors:

Michael Jack Newman - Somerville MA, US
Robert Roth - Newton MA, US
William D. Alexander - Lexington MA, US
Paul van Mulbregt - Wayland MA, US

International Classification:

G10L 15/04
G10L 11/00

US Classification:

704254, 704275, 704E11001, 704E15005

Abstract:

Techniques disclosed herein include systems and methods that enable a voice trigger that wakes-up an electronic device or causes the device to make additional voice commands active, without manual initiation of voice command functionality. In addition, such a voice trigger is dynamically programmable or customizable. A speaker can program or designate a particular phrase as the voice trigger. In general, techniques herein execute a voice-activated wake-up system that operates on a digital signal processor (DSP) or other low-power, secondary processing unit of an electronic device instead of running on a central processing unit (CPU). A speech recognition manager runs two speech recognition systems on an electronic device. The CPU dynamically creates a compact speech system for the DSP. Such a compact system can be continuously run during a standby mode, without quickly exhausting a battery supply.

US Patent:

48052183, Feb 14, 1989

Filed:

Apr 3, 1987

Appl. No.:

7/034842

Inventors:

Paul G. Bamberg - Framingham MA
James K. Baker - West Newton MA
Laurence Gillick - Newton MA
Robert S. Roth - Newtonville MA

Assignee:

Dragon Systems, Inc. - Newton MA

International Classification:

G10L 100

US Classification:

381 43

Abstract:

A method of speech analysis calculates one or more difference parameters for each of a sequence of acoustic frames, where each difference parameter is a function of the difference between an acoustic parameter in one frame and an acoustic parameter in a nearby frame. The method is used in speech recognition which compares the difference parameters of each frame against acoustic models representing speech units, where each speech-unit model has a model of the difference parameters associated with the frames of its speech unit. The difference parameters can be slope parameters or energy difference parameters. Slope parameters are derived by finding the difference between the energy of a given spectral parameter of a given frame and the energy, in a nearby frame, of a spectral parameter associated with a different frequency band. The resulting parameter indicates the extent to which the frequency of energy in the part of the spectrum represented by the given parameter is going up or going down. Energy difference parameters are calculated as a function of the difference between a given spectral parameter in one frame and a spectral parameter in a nearby frame representing the same frequency band.