International Business Machines Corporation

Nuance Communications, Inc.

John Brian Pickering

Brian Hulse

William Bittles

David Granshaw

Jonathan Cook

George Murdoch Clelland

Michael Ken Minakami

Stephen James Hobson

Jose Emir Garza

Matthew D Whitbourne

David R Bell

Steven Powell

Barry Neil Dow

Timothy David Poultney

Benjamin Terrick Staniford

Caroline Edith Maynard

Jeremy Peter James Hughes

Timothy Vincent Pickrell

John Doyle

Fenglian Xu

George M Blue

Michael Cobbett

Gary James Calder

Tim Dunn

Robert Michael Jordan

William Nicholas John Tindall

Andrew James Ritchie

Stephen Graham Lawrence

Paul Andrew Titheridge

Richard M Bicheno

Simon R Gormley

Robert James Mann

Stephen Graham Copinger Lawrence

Jason C Edmeades

Graham Hugh Tuttle

Mark George Stewart Reid Thomas

Ian P Hurworth

Anthony Timothy Farrell

Erhan Mengusoglu

David Laurence Barnes

Paul Reilly

Hoyt McClure Stewart

Andrew McColl Krasun

Philip G Wiloughby

A communication system includes a server and a client that transmits messages to the server. The messages include data and descriptive tags and may be in XML format. The server initiates a negotiation with the client relating to message format switching. If the client indicates that the client can accept message format switching, the server instructs the client to switch further messages to a simpler message format including solely data.

Message switching

A method and system for message handling wherein a structured message is transmitted as a reduced message with the structure removed. The method includes: monitoring messages to be transmitted, the step of monitoring comprising: for each message, identifying one or more elements within the message and identifying the format of content and/or structure encapsulated by those elements; determining the probability that a message will conform to a previously identified format; and using the determined probability to decide when to transmit a format template for the message, wherein the format template is for use by a destination to add the structure removed by a message source back into the message.

Method and system for message handling

An automatic speech recognition (ASR) system includes a speech-responsive application and a recognition engine. The ASR system generates user prompts to elicit certain spoken inputs, and the speech-responsive application performs operations when the spoken inputs are recognized. The recognition engine compares sounds within an input audio signal with phones within an acoustic model, to identify candidate matching phones. A recognition confidence score is calculated for each candidate matching phone, and the confidence scores are used to help identify one or more likely sequences of matching phones that appear to match a word within the grammar of the speech-responsive application. The per-phone confidence scores are evaluated against predefined confidence score criteria (for example, identifying scores below a 'low confidence' threshold) and the results of the evaluation are used to influence subsequent selection of user prompts. One such system uses confidence scores to select prompts for targetted recognition training&#x2014;encouraging input of sounds identified as having low confidence scores. Another system selects prompts to discourage input of sounds that were not easily recognized.

Methods, apparatus and computer programs for automatic speech recognition

A system modeling meta-model language model for a system is extracted from a natural language specification of the system. Syntactic structure is extracted from the specification of a system. The syntactic structure represents a set of at least one syntactic subject. A first mapping is created between a predetermined set of the at least one syntactic subject and respective meta-model elements for a system modeling meta-model language. At least one of the meta-model elements is constructed in accordance with the mapping for each identified syntactic subject. The created meta-model structural elements are created for conversion into a model of the system.

Extracting a system modelling meta-model language model for a system from a natural language specification of the system

Techniques are disclosed for searching a set of documents using search terms. In one embodiment, a summary is provided for each document in the set. Search terms are received, and the set of documents are parsed using the received search terms. A first relevance value is calculated using only the summary of each document. A subset of documents having the highest relevance is provided by using the first relevance value. The subset of documents is parsed using the received search terms, to calculate a second relevance value for each document using the respective document. Query results are provided, the query results including documents having the highest relevance according to the second relevance value.

Querying documents using search terms

A system, method, apparatus, and computer program product uses a plurality of message transmitters to achieve high availability message transmission. The plurality of message transmitters are each operable to transmit a message to a subscriber. In accordance with a determination, for example, of the value of the connections, the message is copied and transmitted to the plurality of the message transmitters so that at least one of the plurality of message transmitters transmits the copy to the subscriber. The at least one message transmitter may be selected by analyzing the message and determining an associated rule based on characteristics of the message.

High availability message transmission

After a client device issues a request, a verification stub component receives the request. Then, the verification stub component obtains a current state of the client device from a client state table. The verification stub component evaluates whether the request is valid or not by analyzing a content and context of the request and checking the request against a protocol matrix. If the request is valid, the verification stub component updates the current state of the client device in the current state table, and forwards the request to a server device. Otherwise, the verification stub component creates a report describing the request is invalid, and evaluates whether the server can process the request. If the server cannot process the request, the verification stub component interrupts the request and forwards the report to the client device. Otherwise, the verification stub component provides the request to the server device.

Method and apparatus to enable protocol verification

A method and system for using input signal quality in an automatic speech recognition system. The method includes measuring the quality of an input signal into a speech recognition system and varying a rejection threshold of the speech recognition system at runtime in dependence on the measurement of the input signal quality. If the measurement of the input signal quality is low, the rejection threshold is reduced and, if the measurement of the input signal quality is high, the rejection threshold is increased. The measurement of the input signal quality may be based on one or more of the measurements of signal-to-noise ratio, loudness, including clipping, and speech signal duration.

Method and system for using input signal quality in speech recognition

A method and apparatus for generating appropriate confirmatory prompts in a speech-enabled, interactive computer system. The method can be incorporated in an interactive voice response system that includes receiving an input audio stream over a voice channel from a users, performing keyword recognition on received input audio as subsequent input audio is being received, and prompting the user with an acknowledgement of the keyword or keywords as subsequent input audio is being received. In another aspect of the method, the volume of the speech input can be continuously monitored. In a further aspect of the method, recognition results and associated confidence values can be combined to select different confirmatory prompts, and the volume is tailored to be the same as, louder than or quieter than the volume of the speech input, so that different types of confirmation can be automatically generated to produce a natural speech-enabled interface.

Computer generated prompting

A method and system for processing a free format short message service (SMS) text message for alerting a mobile phone's user to significant information in the message received by the mobile phone from a sender. Program code stored in memory on the mobile phone executes the method on a processor. A theme, a rheme, at least one structural component, and a sentence type of the normalized message are generated from which a FEATURE FACTOR priority is determined. Semantic content of the normalized message is determined from which an EMOTIONAL FACTOR priority is determined. A TOTAL PRIORITY computed as a sum of the FEATURE FACTOR priority and the EMOTIONAL FACTOR priority is determined to exceed a predetermined threshold, from which it is determined that the received message is significant, which triggers overriding default handset settings of the mobile phone and alerting the user that the received message is significant.

Alerts based on significance of free format text messages

There is disclosed a method and system for preparing a document to be read by a text-to-speech reader. The method can include identifying two or more voice types available to the text-to-speech reader, identifying the text elements within the document, grouping related text elements together, and classifying the text elements according to voice types available to the text-to-speech reader. The method of grouping the related text elements together can include syntactic and intelligent clustering. The classification of text elements can include performing latent semantic analysis on the text elements and characteristics of the available voice types.

Method and apparatus for preparing a document to be read by text-to-speech reader

Growing community of inventors

Winchester, United Kingdom

John Brian Pickering