International Business Machines Corporation

Nuance Communications, Inc.

Jan Sedivy

Dimitri Kanevsky

Stephane Herman Maes

Alexander Zlatsin

Mariusz Sabath

Ponani S Gopalakrishnan

Jan Kleindienst

David Nahamoo

Michael Alan Picheny

Michael Daniel Monkowski

Martin Labsky

Raimo Bakis

Abraham P Ittycheriah

Ladislav Seredi

Paul Andrew Moskowitz

Sara H Basson

Thomas Mario Cipolla

Wlodek Wlodzimierz Zadrozny

Ganesh Nachiappa Ramaswamy

Rafah A Hosn

Mark Edward Epstein

Salim Estephan Roukos

Keith William Grueneberg

Thiruvilwamalai Venkatraman Raman

Daniel E Badt

Arthur J Nadas

Pavel Krbec

Lubos Ures

Radek Hampl

Tomas Beran

Jaroslav Gergic

Vit V Libal

Frantisek Bachleda

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

Clients connecting to a VoiceXML browser obtain a control channel. Using this channel, clients may initialize a new VoiceXML session or attach to an existing VoiceXML session. The client after obtaining a session may perform a range of actions including controlling and monitoring actions.

System, method and architecture for control and multi-modal synchronization of speech browsers

An audio splitting system for sharing speech data associated with the same utterance between multiple speech technologies (consumers). In one aspect, the system comprises one or more queues for storing data, a plurality of consumers each sharing the data stored in the one or more queues and a scheduler for managing the storage of the data in the one or more queues and the consumption of the data in the one or more queues by each of the plurality of consumers. The consumers will register their data requirements and priority requests with the scheduler. The scheduler assigns each of the plurality of consumers to one or more of the queues based on the registered data requirements.

Systems and methods for processing audio using multiple speech technologies

The present invention discloses a method for semantically processing speech for speech recognition purposes. The method can reduce an amount of memory required for a Viterbi search of an N-gram language model having a value of N greater than two and also having at least one embedded grammar that appears in a multiple contexts to a memory size of approximately a bigram model search space with respect to the embedded grammar. The method also reduces needed CPU requirements. Achieved reductions can be accomplished by representing the embedded grammar as a recursive transition network (RTN), where only one instance of the recursive transition network is used for the contexts. Other than the embedded grammars, a Hidden Markov Model (HMM) strategy can be used for the search space.

Enhancement to Viterbi speech processing algorithm for hybrid speech models that conserves memory

A method for generating a language component vocabulary VC for a speech recognition system having a language vocabulary V of a plurality of word forms is disclosed. The method includes: partitioning the language vocabulary V into subsets of word forms based on frequencies of occurrence of the respective word forms; and in at least one of the subsets, splitting word forms having frequencies less than a threshold to thereby generate word form components. Also disclosed is a method for use in speech recognition including: splitting an acoustic vocabulary comprising baseforms into baseform components and storing the baseform components; and, performing sound to spelling mapping on the baseform components so as to generate a baseform components to word parts table for use in subsequent decoding of speech. A method for decoding a speech utterance using language model components and acoustic components, includes the steps of: generating from the utterance a stack of baseform component paths; concatenating baseform components in a path to generate concatenated baseforms, when the concatenated baseform components correspond to a baseform found in an acoustic vocabulary; mapping the concatenated baseforms into words; computing language model (LM) scores associated with the words using a language model, and performing further decoding of the utterance based thereupon.

System and methods for acoustic and language modeling for automatic speech recognition with large vocabularies

Systems and methods for building speech-based applications using reusable dialog components based on VoiceXML (Voice eXtensible Markup Language). VoiceXML reusable dialog components can be used for building a voice interface for use with multi-modal, multi-channel and conversational applications that offer universal access to information anytime, from any location, using any pervasive computing device regardless of its I/O modality. In one embodiment, a framework for reusable dialog components built within the VoiceXML specifications is based on the &#x3c;subdialog&#x3e; tag and ECMAScript parameter objects to pass parameters, configuration and results. This solution is interpreted at the client side (VoiceXML browser). In another embodiment, a framework for reusable dialog components is based on JSP (Java Server Pages) and beans that generate VoiceXML subdialogs. This solution can be evaluated at the server side. These frameworks can be mixed and matched depending on the application.

Reusable voiceXML dialog components, subdialogs and beans

Systems and methods for providing conversational computing in a distributed multi-modal environment and in particular, systems and methods for enabling a user to conversationally communicate with entities in a distributed network using a portable access device, wherein interaction dialog for communicating with entities in the network is described in a manner that is independent of, e.g., user language, access channel(s), I/O modalities of the access device, and regional/international settings, and wherein the user access device and/or a remote server comprise mechanisms for adapting the interaction dialog based on the user language, user preferences, access device characteristics and or regional/international settings, to thereby enable the user to interact with one or more entities in the network.

Systems and methods for providing dialog localization in a distributed environment and enabling conversational communication using generalized user gestures

A system and method that provides data messages to a passive device. A passive device, for example watch, is registered together with the telephone number of a cellular telephone of a subscriber to the data message service. Since the cellular telephone periodically transmits a beacon signal, the wireless network knows its cell location. Accordingly, the system determines the cell location of the cellular telephone and establishes a communication of the subscribed data to the watch via the local cellular provider for the same cell location as that of the subscriber's cellular telephone.

Efficient communication with passive devices

A method and system that permits the purchase of a license to make a limited number of copies of a book. At the time of purchase, the purchaser or user is given a key that contains the ability to obtain the limited number of copies on demand. The key contains a web address that can be used to obtain the authorized copies. In some embodiments, the key is a label in a machine readable form that is readable by a label reader, such as a bar code reader or a magnetic reader. In other embodiments, the key is merely a web address that the user may contact. At the point of sale, the key or record is formed, affixed to the book and also sent to copy tracker. The copy tracker then keeps track of the copies as made and processes each request to make a copy. If permitted, a database is enabled to send an electronic image of the requested copy to the user.

Smart book

System and methods for processing audio using multiple speech technologies

A method and system for forming a virtual network. The method comprises the steps of providing each of a plurality of mobile objects with a transceiver, and transmitting a cellular/radio signals from a source. At least a first of the mobile objects is located in an area where the transceiver of the first mobile object does not receive the signal directly from the source, while a second of the mobile objects is in a position where the transceiver of the second mobile object receives the signal directly from the source. The transceiver of the second mobile object is used to receive the signal directly from the source and to transmit the signal to the transceiver of the first mobile object.

Virtual cooperative network formed by local clients in zones without cellular services

Systems and methods are provided for generating a language component vocabulary VC for a speech recognition system having a language vocabulary V of a plurality of word forms. One method for generating a language component vocabulary VC for a speech recognition system having a language vocabulary V of a plurality of word forms includes partitioning the language vocabulary V into subsets of word forms based on frequencies of occurrence of the respective word forms, in at least one the subsets, splitting word forms having frequencies less than a threshold to thereby generate word form components and generating a language component vocabulary VC including word forms and word form components. The resulting language component vocabulary, which includes word forms and word components, is used to generate a language model that can be efficiently implemented for real-time automatic speech recognition applications for languages with large vocabularies.

Growing community of inventors

Praha, Czechia

Jan Sedivy