Matsushita Electric Industrial Co., Ltd.

Panasonic Corporation

Panasonic Technologies, Inc.

Silencer Devices, Llc.

Panasonic Corporation of North America

Other

Matsushita Electric Corporation of America

Panasonic Intellectual Property Corporation of America

Matsushita Electrical Industrial Co., Ltd.

Matsushita Electronic Industrial Co., Ltd

Jean-Claude Junqua

Roland Kuhn

Luca Rigazio

Patrick Nguyen

Matteo Contolini

Philippe Rene Morin

Robert C Boman

Yi Zhao

David Kryze

Ted H Applebaum

Eugene Seagriff

Tony Davis

Weiying Li

Peter Veprek

Michael Galler

Ricardo Teixeira

Philippe Gelin

John K Howard

Steve D Pearson

Nobuyuki Kunieda

Keiko Morii

Gary David Sasaki

Murat Karaorman

Christophe Cerisara

Florent Perronnin

Younes Souilmi

Yue Fei

Nicholas Kibre

Brian A Hanson

Mike Wang

Kirill Stoimenov

Rathinavelu Chengalvarayan

Steven Pearson

David Koo

Jim Pollock

Dwayne Escola

Raihan Ferdous Sajal

Olivier Thyes

Brice Tsakam

Xavier Anguera Miro

Colin Johnston

Ken Field

Pierluigi Dalla Rosa

Craig Demel

Jeremy Lauraire

Noise abatement within a signal stream containing unwanted signal referred to as noise is performed by acquiring a digitized noise signal and using a digital processor circuit to subdivide the acquired noise signal into different frequency band segments and thereby generate a plurality of segmented noise signals. Then individually for each segmented noise signal, the processor shifts in time the segmented noise signal by an amount dependent on a selected frequency of the segmented noise signal to produce a plurality of shifted segmented noise signals. The precise time shift applied to each noise segment considers the frequency content of the segment and the system processing time. Individually for each segmented noise signal, amplitude scaling is applied. The shifted and amplitude-scaled segmented noise signals are then combined to form a composite anti-noise signal which is output into the signal stream to abate the noise through destructive interference.

Noise cancellation using segmented, frequency-dependent phase cancellation

A noise cancellation technique is presented with improved frequency resolution. The method includes: acquiring a digitized noise signal from an environment in which the audio signal stream is present; receiving a data sample from the digitized noise signal; appending one or more additional samples to the data sample to form a series of samples, where magnitude for each of the one or more additional samples is substantially zero; computing a frequency domain representation of the series of samples in the frequency domain; shifting the frequency domain representation of the series of samples in time using the digital processor circuit, thereby producing a shifted frequency domain representation of the series of samples; converting the shifted frequency domain representation of the series of samples to time domain to form a portion of an anti-noise signal; and outputting the anti-noise signal into the audio signal stream to abate the noise through destructive interference.

Noise cancellation with improved frequency resolution

The patient device, designed for use by patients of reduced cognitive or physical ability, integrates with the network-based portal, designed for use by family members or nursing staff to customize the user experience of the patient. Using the portal the family member pushes user interface customizations as well as content and applications to be run on the patient device. The family member can also control the patient device remotely via the portal. The patient device collects usage data on the set of customizable interface components, and on the pushed content and applications. These data are sent as feedback to the portal, allowing the portal to show the family member which user interface features, content and applications are being used and which are not. Aggregate data collected via the portal generate ranking metrics used to aid in user interface feature selection.

Context dependent application/event activation for people with various cognitive ability levels

The conference room automation apparatus employs a processor-based integrated movement sensor, lights, cameras, and display device, such as a projector, that senses and interprets human movement within the room to control the projector in response to that movement and that captures events occurring in the room. Preferably packed in a common integrated package, the apparatus employs a layered software/hardware architecture that may be readily extended as a platform to support additional third-party functionality.

SmartLight interaction system

The system includes a computer-readable memory having a data structure configured with electronic data indicative of the patient's cognitive ability and/or context information relevant to the patient. The computer system is programmed to dynamically present information based on the patient's cognitive ability and/or context, obtained by accessing the computer-readable memory. The system is thus able to dynamically customize the way the system interacts with the patient, dynamically rendering information and automatically launching specific applications and events based on set parameters to meet the assistive needs of the patient. The system also captures information about the patient's health and wellbeing which is dynamically communicated to the caregiver, which information is also in part based on cognitive ability of the patient and/or context.

The system includes a computer-readable memory having a data structure configured to store information about a time-based event for a patient having reduced cognitive abilities, and optionally also electronic data reflecting the patient's cognitive ability. A networked computer system coupled to the computer-readable memory provides an information communicating interface to the patient. The computer system is programmed to monitor context information relevant to the patient and to dynamically adjust the presentation of the stored information based on the context, and optionally also based upon the patient's cognitive ability.

A call routing and supervising system includes an input receiving customer speech from a remote location, and a voice characteristics extractor extracting voice characteristics from the customer speech, such as language/dialect/accent, age group, gender, and eigendimension coordinates. A customer service representative selector selects one or more customer service representatives based on profiles of the customer service representatives respective of customers having voice characteristics similar to the extracted voice characteristics. In other aspects, a call monitor automatically analyzes dialogue between the customer and the customer service representative, such as detected interruptions, tracked dialogue turns, and recognized key phrases indicating frustration, polity, and/or resolution characteristics of dialogue. The call monitor records performance of the customer service representative respective of customers having the voice characteristics. Automatic call rerouting and/or real-time instruction of call center personnel can also be accomplished based on analysis results.

Intelligent call routing and call supervision method for call centers

Model compression is combined with model compensation. Model compression is needed in embedded ASR to reduce the size and the computational complexity of compressed models. Model-compensation is used to adapt in real-time to changing noise environments. The present invention allows for the design of smaller ASR engines (memory consumption reduced to up to one-sixth) with reduced impact on recognition accuracy and/or robustness to noises.

Block-diagonal covariance joint subspace tying and model compensation for noise robust automatic speech recognition

A noise robustness method operates jointly in a signal domain and a model domain. For example, energy is added in the signal domain for frequency bands where an actual noise level of an incoming signal is lower than a noise level used to train models, thus obtaining a compensated signal. Also, energy is added in the model domain for frequency bands where noise level of the incoming signal or the compensated signal is higher than the noise level used to train the models. Moreover, energy is never removed, thereby avoiding problems of higher sensitivity of energy removal to estimation errors.

Joint signal and model based noise matching noise robustness method for automatic speech recognition

A multilingual text-to-speech system includes a source datastore of primary source parameters providing information about a speaker of a primary language. A plurality of primary filter parameters provides information about sounds in the primary language. A plurality of secondary filter parameters provides information about sounds in a secondary language. One or more secondary filter parameters is normalized to the primary filter parameters and mapped to a primary source parameter.

Multilingual text-to-speech system with limited resources

A media capture device has an audio input receptive of user speech relating to a media capture activity in close temporal relation to the media capture activity. A plurality of focused speech recognition lexica respectively relating to media capture activities are stored on the device, and a speech recognizer recognizes the user speech based on a selected one of the focused speech recognition lexica. A media tagger tags captured media with generated speech recognition text, and a media annotator annotates the captured media with a sample of the user speech that is suitable for input to a speech recognizer. Tagging and annotating are based on close temporal relation between receipt of the user speech and capture of the captured media. Annotations may be converted to tags during post processing, employed to edit a lexicon using letter-to-sound rules and spelled word input, or matched directly to speech to retrieve captured media.

Voice tagging, voice annotation, and speech recognition for portable devices with optional post processing

A portable device increases user access to equipment utilizing a communications interface providing communication with the equipment in accordance with various, combinable embodiments. In one embodiment, a speech generator generates speech based on commands relating to equipment operation, which may be received from the equipment via the communications interface. A selection mechanism allows the user to select commands and thereby operate the equipment. In another embodiment, a command navigator navigates commands based on user input by shifting focus between commands, communicates a command having the focus to the speech generator, and allows the user to select a command. In a further embodiment, a phoneticizer converts the commands and/or predetermined navigation and selection options into a dynamic speech lexicon, and a speech recognizer uses the lexicon to recognize a user navigation input and/or user selection of a command. Speaker verification can also be used to enhance security using a speech biometric.

Portable device for enhanced security and accessibility

A speaker authentication system includes a data fuser operable to fuse voiceprint match attempt results with additional information to assist in authenticating a speaker providing audio input. In other aspects, the system includes a data store of speaker voiceprints and a voiceprint matching module adapted to receive an audio input and operable to attempt to assist in authenticating a speaker by matching the audio input to at least one of the speaker voiceprints. The voiceprint matching module adjusts a confidence of voiceprint match attempt results by at least one of: (a) a number of utterance repetitions upon which a matching speaker voiceprint has been trained; or (b) a passage of time since a training occurrence associated with a matching speaker voiceprint.

Speaker authentication by fusion of voiceprint match attempt results with additional information

Growing community of inventors

Santa Barbara, CA, United States of America

Jean-Claude Junqua