International Business Machines Corporation

Nuance Communications, Inc.

Andrew Aaron

Jonathan Lenchner

Maharaj Mukherjee

Shang Qing Guo

Ellen E Kislal

Mark Edward Podlaseck

Hyun Kyu Seo

David O S Melville

John C Nelson

Dimitri Kanevsky

Edward Emile Kelley

Hariklia Deligianni

Bhuvana Ramabhadran

Joel W Branch

Ellen Marie Eide

Wael Mohamed Hamza

Daniel A Mazzella

David Angelo Ferrucci

Michael Alan Picheny

David M Lubensky

Maria Elena Smith

Raimo Bakis

Dario Gil

Kenneth Blair Ocheltree

John Ferdinand Pitrelli

Zhi Wei Shuang

Charles T Rutherfoord

Subrata Kumar Das

Emrah Akin Sisbot

A mistranscription generated by a speech recognition system is identified. A received utterance is matched to a first utterance member within a set of known utterance members. The matching operation matches fewer than the first plural number of words in the received utterance and the received utterance varies in a first particular manner as compared to a first word in a first slot in the first utterance member. The received utterance is sent to a mistranscription analyzer component which increments evidence that the received utterance is evidence of a mistranscription. Once the incremented evidence for the mistranscription exceeds a threshold, future received utterances containing the mistranscription are treated as though the first word was recognized.

Learning transcription errors in speech recognition tasks

A method controls a robot. One or more processors receive sensor readings from one or more sensors that are monitoring a human in real time, where the human is currently observing a robotic action by a robot, and where the robotic action is a physical movement performed by the robot. The processor(s) determine a cognitive state of the human while the human is observing the robotic action by the robot, and then adjust the robotic action being performed by the robot based on the cognitive state of the human.

Robotic physical movement adjustment based on user emotional state

A touch detection system is provided. The touch detection system includes a hardware processor. The touch detection system further includes a touch input surface mounted using a set of torque sensors, where a location of a touch on the touch input surface is calculated by the hardware processor based on collective results from the set of torque sensors, the touch comprising a user input.

Touch sensitivity for robotically operated displays

A system, computer program product and computer-implemented method for presenting data. The system includes a sensor that obtains a measurement related to a viewing parameter of a viewer in the environment, a presentation surface that provides data for viewing by the viewer, and a processor. The viewing parameter of the viewer indicates the viewer's ability to perceive the data. The processor determines the viewing parameter of the viewer from the obtained measurement and dynamically alters a presentation parameter of the presentation surface based at least in part on the viewing parameter of the viewer.

Adaptive display environment for dynamic applications

A mistranscription generated by a speech recognition system is identified. A first class of utterance members is provided for use by the speech recognition system, each utterance class member consisting of a respective number of words. The first class is defined by a first common meaning and a first common system response if a class member of the first class is recognized. If the speech recognition system matches a received utterance to a first class member of the first class, the received utterance is sent to a mistranscription analyzer. The received utterance contains a mistranscription as compared to the first class member. Evidence is incremented by the mistranscription analyzer that the received utterance is evidence of the mistranscription of the first class member. If the incremented evidence for the mistranscription of the first class member exceeds a first threshold, a second class member is added to the first class of utterance members based on the mistranscription of the first class member. If the speech recognition system recognizes a second received utterance which matches the second class member, performing the common system response.

Class based learning for transcription errors in speech recognition tasks

A mistranscription generated by a speech recognition system is identified. A set of known utterance members is provided for use by a speech recognition system. Each utterance member consists of a respective plural number of words. A received utterance is matched to a first utterance member within the set of known utterance members. The first utterance member is a closest matching utterance member and has a first plural number of words. The matching operation matches fewer than the first plural number of words in the received utterance and the received utterance varies in a first particular manner as compared to a first word in a first slot in the first utterance member. The received utterance is sent to a mistranscription analyzer component which increments evidence that the received utterance is evidence of a mistranscription. Once the incremented evidence for the mistranscription exceeds a threshold, future received utterances containing the mistranscription are treated as though the first word was recognized.

A computer program product is provided for improving comprehension in rapid serial visual presentation. The product includes a non-transitory computer readable storage medium having program instructions embodied therewith executable by a computer to cause the computer to perform a method. The method includes determining, by a cognitive load estimator, a cognitive load of a plurality of words included in a rapid serial visual presentation by using at least one metric. The cognitive load is determined on any of a word level and a word sequence level. The method includes calculating, by a word presentation rate calculator, a variable presentation rate for the words based on the cognitive load. The method includes controlling, by a rate controller, a displaying of the words on a display in accordance with the calculated variable presentation rate. The rate controller temporarily reduces the variable presentation rate responsive to the cognitive load being above a threshold.

Comprehension in rapid serial visual presentation

Methods, systems, and computer program products for normalizing the speaking volume of participants in meetings are provided herein. A computer-implemented method includes collecting speaking volume samples from multiple participants of a group event, wherein said speaking volume samples are derived from microphones; comparing said speaking volume samples to a predetermined volume level; determining which of the microphones is being used by which of the multiple participants during the group event; automatically amplifying a broadcast volume of each audio output device during periods of audio input attributed to each microphone determined as being used by the participants attributed to a speaking volume sample that is below the predetermined volume level; and automatically reducing a broadcast volume of each audio output device during periods of audio input attributed to each microphone determined as being used by the participants attributed to a speaking volume sample that exceeds the predetermined volume level.

Normalizing the speaking volume of participants in meetings

A method for highlighting an electronic document comprises mining one or more sources to obtain data corresponding to a user, inferring one or more interests of the user based on the data, analyzing the electronic document to determine at least one portion of the electronic document related to the one or more interests, and automatically highlighting the at least one portion of the electronic document related to the one or more interests.

Personalized highlighter for textual media

A private message can be delivered by a computer system to a user in an environment, using a display in the environment, a user device, and sensors in the space. The device is detected and connected to a system once it enters a space. A private message is directed to the device and it is received by the system, which notifies the device that the message has been received. The user can then look up at a device to display the message and the system can determine where the user is looking, in order to deliver the message using rapid serial visual presentation to the user on the display device.

Rapid serial visual presentation to deliver messages in social spaces

Computer-implemented methods are provided. A method includes determining, by a processor, a cognitive load of a plurality of words included in a rapid serial visual presentation by using at least one metric. The cognitive load is determined on any of a word level and a word sequence level. The method further includes calculating, by the processor, a variable presentation rate for the plurality of words based on the cognitive load. The method also includes controlling, by the processor, a displaying of the plurality of words on a display device in accordance with the calculated variable presentation rate based on a threshold by temporarily reducing the variable presentation rate responsive to the cognitive load being above the threshold.

Methods and a system are provided. A method includes receiving a plurality of words comprised in a Rapid Serial Visual Presentation. The method further includes determining a cognitive load of the plurality of words by using at least one metric. The cognitive load is determined on any of a word level and a word sequence level. The method also includes calculating a variable presentation rate for the plurality of words based on the cognitive load. The variable presentation rate is capable of being varied on any of the word level and the word sequence level. The method additionally includes controlling a displaying of the plurality of words on a display device in accordance with the calculated variable presentation rate.

Aspects of the disclosure include managing a set of data associated with a corpus. By analyzing the corpus, a domain is established to characterize the subject matter of the set of data. A user identifier is generated for a portion of the set of data. Based upon a credibility computation, a quality factor for a portion of the set of data is determined. The credibility computation includes using both the domain and the user identifier to determine the quality factor for the portion of the set of data. The quality factor for the portion of the set of data is compared with a threshold. In response to a quality factor for a portion of the set of data exceeding the threshold, the portion of the set of data is selected.

Managing a set of data

A method for improving the performance of a noise cancellation device, the method includes determining whether one or more noise making objects (NMO) are near an audible range of the noise cancellation device and receiving a signal from the one or more NMOs indicative of a kind of noise the one or more NMOs is generating. The method also includes selecting a specific noise cancellation model to reduce an expected noise in response to the received kind of noise the one or more NMOs is generating.

Growing community of inventors

Ardsley, NY, United States of America

Andrew Aaron