International Business Machines Corporation

Nuance Communications, Inc.

Cerence Operating Company

Jan Kleindienst

Martin Labsky

Tomas Macek

Ganesh Nachiappa Ramaswamy

Jan Curin

Daniel Mark Coffman

Ladislav Kunc

Holger Quast

Stephane Herman Maes

David Nahamoo

Jan Sedivy

Rafah A Hosn

Thiruvilwamalai Venkatraman Raman

Tomás Macek

Ladislav Seredi

Popani Gopalakrishnan

Martin Labský

Jan Vystrcil

William F Ganong, Iii

Ponani S Gopalakrishnan

Garrett Weinberg

Lars Koenig

Victor S Moore

Rahul P Akolkar

Chalapathy Venkata Neti

Nils Lenke

Keith William Grueneberg

Charles Francis Wiecha

Simona Gandrabur

Eric Buist

Jiri Havelka

Martin Schmid

Bart D'hoore

Lubos Ures

David R Boloker

Jaroslav Gergic

Photina Jaeyun Jang

Kyle A Jamieson

Lars Köenig

Arthur Kantor

Frantisek Bachleda

Rafah A c/o Ibm United Kingdom Limitied Hosn

Stephane H c/o Ibm Uk Ltd IntelPropLaw Maes

According to some aspects, a method of searching for content in response to a user voice query is provided. The method may comprise receiving the user voice query, performing speech recognition to generate N best speech recognition results comprising a first speech recognition result, performing a supervised search of at least one content repository to identify one or more supervised search results using one or more classifiers that classify the first speech recognition result into at least one class that identifies previously classified content in the at least one content repository, performing an unsupervised search of the at least one content repository to identify one or more unsupervised search results, wherein performing the unsupervised search comprises performing a word search of the at least one content repository, and generating combined results from among the one or more supervised search results and the one or more unsupervised search results.

Systems and methods for speech-based searching of content repositories

Methods, systems, computer-readable media, and apparatuses for providing search disambiguation using contextual information and domain ontologies are presented. In some embodiments, a computing device may receive a natural language input from a user. The computing device may identify a plurality of hypotheses for the natural language input. The computing device may map the plurality of hypotheses to one or more concepts of a plurality of concepts of an ontology by annotating the one or more concepts. The ontology may include the plurality of concepts respectively connected by a plurality of relations. The computing device may determine that there is an imperfect match between the annotated one or more concepts and annotations of answers. In response, the computing device may disambiguate the annotated one or more concepts using the ontology. The computing device may present output to the user based on the disambiguation.

Contextual search disambiguation

Techniques for error correction using a history list comprising at least one misrecognition and correction information associated with each of the at least one misrecognitions indicating how a user corrected the associated misrecognition. The techniques include converting data input from a user to generate a text segment, determining whether at least a portion of the text segment appears in the history list as one of the at least one misrecognitions, if the at least a portion of the text segment appears in the history list as one of the at least one misrecognitions, obtaining the correction information associated with the at least one misrecognition, and correcting the at least a portion of the text segment based, at least in part, on the correction information.

Methods and apparatus for correcting recognition errors

Aspects described herein provide solutions to problems posed by a user. Input that includes a specified subject may be received from a user. A specified descriptor for the specified subject may be obtained from a hierarchical taxonomy associated with the specified subject. An information repository may be searched based on the specific descriptor, and entries stored at the information repository that are associated with the specific descriptor may be indicated in a list of results. The specific descriptor may be iteratively generalized to obtain generalized descriptors, and the information repository may be searched based on the generalized descriptors. The generalized descriptors may also be specialized to obtain specialized descriptors, and the information repository may be searched based on the specialized descriptors. The list of results may include entries stored at the information repository that are respectively associated with the generalized descriptors and the specialized descriptors.

Recommendation system with hierarchical mapping and imperfect matching

An automotive text display arrangement is described which includes a driver text display positioned directly in front of an automobile driver and displaying a limited amount of text to the driver without impairing forward visual attention of the driver. The arrangement may include a boundary insertion mode wherein when the active text position is an active text boundary, new text is inserted between the text items separated by the active text boundary, and when the active text position is an active text item, new text replaces the active text item. In addition or alternatively, there may be a multifunctional text control knob offering multiple different user movements, each performing an associated text processing function.

Text browsing, editing and correction methods for automotive applications

Methods and apparatus to receive a first data stream, such as a public broadcast, and receive a second data stream, such as a private data stream, containing emails, for example. A user profile can be used to generate a data output stream for the user from the first and second data streams. The user profile can contain preferences for segments within the first and second data streams.

Methods and apparatus for providing mixed audio streams

Methods, devices, and computer program products for recognizing and responding to natural language input are described herein. Natural language input is received at a natural language input interface of a computing device and transformed into computer-usable text. A natural language input recognizer obtains evidence from one or more evidence source and generates an evidence graph based on the evidence obtained. Evidence may be obtained asynchronously, and the natural language input recognizer may update the evidence graph upon receipt of additional evidence. The natural language input recognizer generates a set of recognition hypotheses based on the evidence graph and selects one of the recognition hypotheses as a recognition result for the natural language input. Semantic models, evidence models, and response models may be employed to generate the evidence graph and respond to the recognition result selected for the natural language input.

Evidence-based natural language input recognition

A mechanism is provided for generating a natural language question for a given input text. The input text is parsed using a minimal recursion semantics (MRS) generating grammar to obtain a minimal recursion semantics (MRS) representation of the input text. Semantic rote labelling transforms the input text into at least one semantic association of a verb and semantic arguments of the verb, the semantic arguments of the verb being fragments of the input text. A question type is received for at least one verb/semantic argument association. The MRS representation of the input text is transformed into a MRS representation of one or more questions based on the at least one semantic association of the verb and respective question types. At least one question of the one or more questions is generated based on the MRS representation of the at least one question using the MRS generating grammar.

Automatic question generation from natural text

A textual message processing system and method are described for use in a mobile environment. A user messaging application processes at least one user textual message during a user messaging session. A semantic annotation module identifies one or more semantically salient terms in the user textual message, and annotates the user textual message with annotation terms having a low semantic distance to the semantically salient terms. A user message history stores the annotated textual messages. The semantic annotation module may further annotate the user textual message with situational meta-data characterizing the user textual message. There may be a message search module for using one or more keywords to search the user message history including the annotation terms, and identifying as a search match any annotated textual messages within a semantic distance threshold of the one or more keywords.

Concept search and semantic annotation for mobile messaging

Systems and methods for multi-modal messaging that enable a user to compose, send and retrieve messages, such as SMS, MMS, IM or ordinary e-mail messages, for example, using one or more I/O (input/output) modalities (e.g., speech I/O and/or GUI I/O). A method for composing messages combines the advantages of a multi-modal interface (e.g., grammar-based speech and touchscreen or similar input devices) and message templates, which allows a user to construct a message with significantly less effort in a fraction of the time required by conventional methods. The user can dictate his/her messages using speech and/or GUI input, for example, based on a library of message templates which can be personalized by the user to fit his/her social interaction needs.

Multi-modal messaging

Techniques disclosed herein include systems and methods for managing user interface responses to user input including spoken queries and commands. This includes providing incremental user interface (UI) response based on multiple recognition results about user input that are received with different delays. Such techniques include providing an initial response to a user at an early time, before remote recognition results are available. Systems herein can respond incrementally by initiating an initial UI response based on first recognition results, and then modify the initial UI response after receiving secondary recognition results. Since an initial response begins immediately, instead of waiting for results from all recognizers, it reduces the perceived delay by the user before complete results get rendered to the user.

Method to provide incremental UI response based on multiple asynchronous evidence about user input

A method for providing a voice application includes executing control flow logic modeling a dialog flow with a user via a voice browser. The control flow logic produces a disambiguation requirement. A disambiguation module is initiated and a set of at least two candidates and partitioning criteria is sent from the control flow logic to the module. Attributes of the candidates are analyzed to determine a partitioning score for each attribute indicative of ability to distinguish between candidates based on the partitioning criteria. The attributes are sorted based on the partitioning scores. The user is queried based on a top-sorted attribute and results of the query are used to reduce the set of candidates. The steps of analyzing, sorting, and querying are repeated until the set of candidates is reduced to a single candidate. The single candidate is returned to the control flow logic for continued execution.

Automated confirmation and disambiguation modules in voice applications

Techniques for presenting data input as a plurality of data chunks including a first data chunk and a second data chunk. The techniques include converting the plurality of data chunks to a textual representation comprising a plurality of text chunks including a first text chunk corresponding to the first data chunk and a second text chunk corresponding to the second data chunk, respectively, and providing a presentation of at least part of the textual representation such that the first text chunk is presented differently than the second text chunk to, when presented, assist a user in proofing the textual representation.

Methods and apparatus for proofing of a text input

An automatic conversational system has multiple computer-implemented dialogue components for conducting an automated dialogue process with a human user. A user client delivers dialogue output prompts to the human user and receives dialogue input responses from the human user including speech inputs. An automatic speech recognition engine processes the speech inputs to determine corresponding sequences of representative text words. A natural language understanding (NLU) processing arrangement processes the dialogue input responses and the text words to determine corresponding semantic interpretations. The NLU processing arrangement includes an anaphora processor that accesses different information sources characterizing dialogue context, linguistic features, and NLU features to identify unresolved anaphora in the text words needing resolution in order to determine a semantic interpretation. A dialogue manager manages the dialogue process with the human user based on the semantic interpretations.

Anaphora resolution using linguisitic cues, dialogue context, and general knowledge

Growing community of inventors

Nove Straseci, Czechia

Jan Kleindienst