Abstract: Techniques are disclosed for automatically generating structured documents based on speech, including identification of relevant concepts and their interpretation. In one embodiment, a structured document generator uses an integrated process to generate a structured textual document (such as a structured textual medical report) based on a spoken audio stream. The spoken audio stream may be recognized using a language model which includes a plurality of sub-models arranged in a hierarchical structure. Each of the sub-models may correspond to a concept that is expected to appear in the spoken audio stream. Different portions of the spoken audio stream may be recognized using different sub-models. The resulting structured textual document may have a hierarchical structure that corresponds to the hierarchical structure of the language sub-models that were used to generate the structured textual document.

Abstract: In one example, a phrase analyzer may analyze a text input stream to identify phrases contained in the text input stream. The phrase analyzer may receive a specification, which includes dictionaries of phrases and synonyms, and a specification of the phrases, or sequences of phrases to be matched. The phrase analyzer may compare the input stream to the specification and may produce, as output, an identification of which phrases appear in the input stream, and where in the input stream those phrases occur.

Abstract: A method and system for determining a lexical association of phrasal terms are described. A corpus having a plurality of words is received, and a plurality of contexts including one or more context words proximate to a word in the corpus is determined. An occurrence count for each context is determined, and a global rank is assigned based on the occurrence count. Similarly, a number of occurrences of a word being used in a context is determined, and a local rank is assigned to the word-context pair based on the number of occurrences. A rank ratio is then determined for each word-context pair. A rank ratio is equal to the global rank divided by the local rank for a word-context pair. A mutual rank ratio is determined by multiplying the rank ratios corresponding to a phrase. The mutual rank ratio is used to identify phrasal terms in the corpus.

Abstract: A speech recognition device includes one intention extracting language model and more in which an intention of a focused specific task is inherent, an absorbing language model in which any intention of the task is not inherent, a language score calculating section that calculates a language score indicating a linguistic similarity between each of the intention extracting language model and the absorbing language model, and the content of an utterance, and a decoder that estimates an intention in the content of an utterance based on a language score of each of the language models calculated by the language score calculating section.

Abstract: Data defining an avatar is received, over a network. The data defining the avatar is stored to a computer readable medium. A sprite sheet comprising a plurality of sprites is created, using at least one computing device, using the data defining the avatar. Each sprite comprises a partial rendering of the respective avatar and at least one run-time parameter comprising a sprite attribute. A plurality of requests are received, over the network, for the avatar from a plurality of user applications. The data defining the avatar and the sprite sheet are transmitted, over the network, to each of the requesting games, whereby each respective user application is enabled to display the sprites in the sprite sheet. The user application is enabled to set the run-time parameter associated with each of the sprites in the sprite sheet such that each respective sprite is thereby customized to the application.

Abstract: A speech understanding apparatus includes a speech recognition unit which performs speech recognition of an utterance using multiple language models, and outputs multiple speech recognition results obtained by the speech recognition, a language understanding unit which uses multiple language understanding models to perform language understanding for each of the multiple speech recognition results output from the speech recognition unit, and outputs multiple speech understanding results obtained from the language understanding, and an integrating unit which calculates, based on values representing features of the speech understanding results, utterance batch confidences that numerically express accuracy of the speech understanding results for each of the multiple speech understanding results output from the language understanding unit, and selects one of the speech understanding results with a highest utterance batch confidence among the calculated utterance batch confidences.

Abstract: Systems and methods for processing speech are provided. A system may include a speech recognition interface and a processor. The processor may convert speech received from a call at the speech recognition interface to at least one word string. The processor may parse each word string of the at least one word string into first objects and first actions. The processor may access a synonym table to determine second objects and second actions based on the first objects and the first actions. The processor may also select a preferred object and a preferred action from the second objects and the second actions.

Abstract: A method is disclosed for quantitatively assessing information in natural language contents related to an object name. The method includes identifying a sentence in a document, determining a subject and a predicate in the sentence, and retrieving an object-specific data set related to the object name. The object-specific data set includes property names and association-strength values. Each property name is associated with an association-strength value. The method also includes identifying a first property name in the property names that matches the subject, assigning a first association-strength value associated with the first property name to the subject, identifying a second property name in the property names that matches the predicate, assigning a second association-strength value associated with the second property name to the predicate, and multiplying the first association-strength value and the second association-strength value to produce a sentence information index.

Abstract: A data processing apparatus includes a speech recognition unit configured to perform continuous speech recognition on speech data, a related word acquiring unit configured to acquire a word related to at least one word obtained through the continuous speech recognition as a related word that is related to content corresponding to content data including the speech data, and a speech retrieval unit configured to retrieve an utterance of the related word from the speech data so as to acquire the related word whose utterance has been retrieved as metadata for the content.

Abstract: A message processing system is a server (1) incorporating software functions (15) which operate without user intervention to intercept (16, 17) SMS text messages, to translate the original text content in the short message body to compress it, and to forward the converted message on in the network. The content compression frees up space, and some or all of this space may be taken up by addition of tagging information such as text tags or sponsored or advertising content or user data header (UDH) information elements. The translator is part of a message converter (15) and parses the content and identifies candidate words for translation, and then iterates over the content, replacing the candidate word to allow the addition of one or more text tags to the message. The translation could be performed on just enough words and phrases to free up the required space, or it could be on the message as a unit, so as it give it uniformity of presentation.

Abstract: In one embodiment, an apparatus for automated generation of subject line content for e-mail messages includes an input operable to receive content data including text-based information corresponding to a body of an e-mail message, a text analyzer including logic operable to analyze received content data, a topic extractor including logic operable to extract topic data in accordance with an output of the text analyzer, a string generator including logic operable to generate subject line data in accordance with an output of the topic extractor, and a message output operable to output a multi-field e-mail message having a body field inclusive of the content data and a subject line field inclusive of generated subject line data.

Abstract: A system and method of developing rules for text processing enable retrieval of instances of named entities in a predetermined semantic relation (such as the DATE and PLACE of an EVENT) by extracting patterns from text strings in which attested examples of named entities satisfying the semantic relation occur. The patterns are generalized to form rules which can be added to the existing rules of a syntactic parser and subsequently applied to text to find candidate instances of other named entities in the predetermined semantic relation.

Abstract: Non-verbalized tokens, such as punctuation, are automatically predicted and inserted into a transcription of speech in which the tokens were not explicitly verbalized. Token prediction may be integrated with speech decoding, rather than performed as a post-process to speech decoding.

Abstract: An enhanced system for speech interpretation is provided. The system may include receiving a user verbalization and generating one or more preliminary interpretations of the verbalization by identifying one or more phonemes in the verbalization. An acoustic grammar may be used to map the phonemes to syllables or words, and the acoustic grammar may include one or more linking elements to reduce a search space associated with the grammar. The preliminary interpretations may be subject to various post-processing techniques to sharpen accuracy of the preliminary interpretation. A heuristic model may assign weights to various parameters based on a context, a user profile, or other domain knowledge. A probable interpretation may be identified based on a confidence score for each of a set of candidate interpretations generated by the heuristic model. The model may be augmented or updated based on various information associated with the interpretation of the verbalization.

Abstract: At least one transaction and at least one transaction parameter that is allocated thereto are determined based on at least one user statement in order to trigger at least one first and second background application via a universal language dialogue system, first transactions and first transaction parameters being assigned to the first background application and second transactions and second transaction parameters being associated with the second background application. The first and second transactions as well as the first and second transaction parameters are linked together via a universal dialogue specification which is evaluated to determine the at least one transaction and at least on associated transaction parameter in order to trigger at least one of the background application via the universal language dialogue system.

Abstract: The present invention relates to a method and apparatus of translating a language using voice recognition. The present invention provides a method of translating a language using voice recognition, comprising: receiving a voice input comprising a first language; acquiring at least one recognition candidate corresponding to the voice input by performing voice recognition on the voice input; providing a user interface for selecting at least one of the acquired at least one recognition candidate; and outputting a second language corresponding to the selected at least one recognition candidate, wherein the type of the user interface is determined according to the number of the acquired at least one recognition candidate, and an apparatus of translating a language using voice recognition for implementing the above method.

Abstract: Disclosed is a speech recognition system in which a common data processing means performs speech recognition of a speech captured by a speech input means to generate recognition result hypotheses which is not biased to one of applications and an adaptation data processing means regenerates recognition result hypotheses, using adaptation data and adaptation processing for each application. The adaptation data processing means provides to each application the recognition result recalculated for each application.

Abstract: A system and method are provided for receiving speech and/or non-speech communications of natural language questions and/or commands and executing the questions and/or commands. The invention provides a conversational human-machine interface that includes a conversational speech analyzer, a general cognitive model, an environmental model, and a personalized cognitive model to determine context, domain knowledge, and invoke prior information to interpret a spoken utterance or a received non-spoken message. The system and method creates, stores and uses extensive personal profile information for each user, thereby improving the reliability of determining the context of the speech or non-speech communication and presenting the expected results for a particular question or command.

Abstract: An automated method is described for developing an automated speech input semantic classification system such as a call routing system. A set of semantic classifications is defined for classification of input speech utterances, where each semantic classification represents a specific semantic classification of the speech input. The semantic classification system is trained from training data having little or no in-domain manually transcribed training data, and then operated to assign input speech utterances to the defined semantic classifications. Adaptation training data based on input speech utterances is collected with manually assigned semantic labels. When the adaptation training data satisfies a pre-determined adaptation criteria, the semantic classification system is automatically retrained based on the adaptation training data.

Abstract: Embodiments for the conversion of Computational Independent Model (CIM) rule expressions into semantically non-ambiguous syntax trees are disclosed. In accordance with one embodiment, a method includes analyzing a sentential structure of a Computational Independent Model (CIM) rule expression for clauses. The clauses include at least one expression and at least one rule. The method further includes constructing a semantically non-ambiguous LF syntax tree from the CIM rule expression. The construction being implemented using a logical form (LF) model.

Abstract: An apparatus may include a processor configured to receive vocabulary entry data. The processor may be further configured to determine a class for the received vocabulary entry data. The processor may be additionally configured to identify one or more languages for the vocabulary entry data based upon the determined class. The processor may also be configured to generate a phoneme sequence for the vocabulary entry data for each identified language. Corresponding methods and computer program products are also provided.

Abstract: A semantic conversion system (1900) includes a self-learning tool (1902). The self-learning tool (1902) receives input files from legacy data systems (1904). The self-learning tool (1902) includes a conversion processor (1914) that can calculate probabilities associated with candidate conversion terms so as to select an appropriate conversion term. The self-learning tool (1902) provides a fully attributed and normalized data set (1908).

Abstract: A speech recognition apparatus capable of attaining high recognition accuracy within practical processing time using a computing machine having standard performance by appropriately adapting a language model to a speech about a certain topic, irrespectively of a degree of detail and diversity of the topic and irrespectively of a confidence score of an initial speech recognition result is provided.

Abstract: Word lattices that are generated by an automatic speech recognition system are used to generate a modified word lattice that is usable by a spoken language understanding module. In one embodiment, the spoken language understanding module determines a set of salient phrases by calculating an intersection of the modified word lattice, which is optionally preprocessed, and a finite state machine that includes a plurality of salient grammar fragments.

Abstract: Human-like response emulator stores a library (14) comprising one or more different subject matter data structures. Each data structure comprising a set of stimuli related to the subject matter of the data structure and one or more output instructions associated with each stimulus. Each output instruction is for producing a human-like response to the associated stimulus. The emulator receives a stimulus (16, 18). The emulator looks up output instructions in each data structure that are associated with the received stimulus.

Abstract: The invention relates to a task classification system (900) that interacts with a user. The task classification system (900) may include a recognizer (920) that may recognize symbols in the user's input communication, and a natural language understanding unit (900) that may determine whether the user's input communication can be understood. If the user's input communication can be understood, the natural language understanding unit (930) may generate understanding data. The system may also include a communicative goal generator that may generate communicative goals based on the symbols recognized by the recognizer (920) and understanding data from the natural language understanding unit (930). The generated communicative goals may be related to information needed to be obtained from the user.

Abstract: A method and system for labeling a selected word of a sentence using a deep neural network includes, in one exemplary embodiment, determining an index term corresponding to each feature of the word, transforming the index term or terms of the word into a vector, and predicting a label for the word using the vector. The method and system, in another exemplary embodiment, includes determining, for each word in the sentence, an index term corresponding to each feature of the word, transforming the index term or terms of each word in the sentence into a vector, applying a convolution operation to the vector of the selected word and at least one of the vectors of the other words in the sentence, to transform the vectors into a matrix of vectors, each of the vectors in the matrix including a plurality of row values, constructing a single vector from the vectors in the matrix, and predicting a label for the selected word using the single vector.

Abstract: A speech recognition device and a method thereof are adapted to recognize a Chinese word. The speech recognition device includes a lexicon model, a language model, a speech recognition module, and a parsing module. The lexicon model keeps a plurality of words. The speech recognition module performs a speech recognition processing on a voice signal conforming to a syntax structure of Chinese word description. The speech recognition processing searches words related to the Chinese word description from the lexicon model according to a feature of the Chinese word description, and produces a literal word series in digital data form by referring a syntax combination probability. The language model based on the syntax structure of Chinese word description provides the syntax combination probability according to combination relations between the searched words. The parsing module analyzes the syntax structure of the literal word series for retrieving the Chinese word.

Abstract: A system for idiom simultaneous translation applied to telephonic equipment, conventional or mobile phones and service rendered by a telephonic company allowing complete understanding of a conversation between speaker and listener agents, when both speak different idioms, with the main objective is to allow a concurrent translation of the native idiom of the speaker agent to the native idiom of the listener agent, through the introduction of a idiom translation system, composed by a computer readable program supported by a hardware and linked to a voice recognition device, which may be integrated to a generic telephonic equipment, both conventional and mobile, when in a local application, or directly installed in a telephonic central system, as a service to be rendered to its subscribers.

Abstract: A speech recognition method including for a spoken expression: a) providing a vocabulary of words including predetermined subsets of words, b) assigning to each word of at least one subset an individual score as a function of the value of a criterion of the acoustic resemblance of that word to a portion of the spoken expression, c) for a plurality of subsets, assigning to each subset of the plurality of subsets a composite score corresponding to a sum of the individual scores of the words of said subset, d) determining at least one preferred subset having the highest composite score.

Abstract: In one embodiment, a set of target search terms for a search is received. Candidate terms are selected, where a candidate term is selected to reduce an ontology space of the search. The candidate terms are to a computer to recommend the candidate terms as search terms. In another embodiment, a document stored in one or more tangible media is accessed. A set of target tags for the document is received. Terms are selected, where a term is selected to reduce an ontology space of the document. The terms are sent to a computer to recommend the terms as tags.

Abstract: Provided are an apparatus and method for post-processing a dialogue error in a speech dialogue system using multilevel verification, in which both of a user's current utterance and a whole dialogue flow are taken into account through the multilevel verification including speech recognition results analysis, linguistic analysis, discourse analysis and dialogue analysis. As a result, various errors that may occur in the speech dialogue system are detected, and error post-processing appropriate to a detected error type is performed, so that speech recognition errors may be reduced.

Abstract: A device (100) for automatically creating an information analysis report includes a processing device (1), an input device (2), a recording device (3), and an output device (4). When creating an information analysis report: a document to be surveyed and a document to be compared are specified and inputted; an information analysis condition is inputted; a population document formed by a document group similar to the document to be surveyed is selected from the document to be compared; an index word characteristic to the population document of the document to be surveyed is extracted; according to the population document and the index word, an information analysis report representing the feature of the document to be surveyed is created; and the created information report document is outputted to display means, recording means, or communication means.

Abstract: Disclosed is a method to perform natural language (NL) processing. The method includes accessing a data source having one or more data portions, and applying multi-stage NL processing on the one or more data portions, using a dynamically generated set of concepts relating to one or more subject matters and relationships between at least some of the concepts, to determine the association of the one or more data portions with one or more of the concepts.

Abstract: Disclosed are a method (500), apparatus (100) and computer program product for generating a mixed-initiative dialog to obtain information for dialog slots. A composite grammar dependent upon a set of unfilled slots is constructed (501). A prompt, dependent upon the a set of unfilled slots, is presented (309) to a user. An utterance is received (301) from the user in response to said prompt. Relevant information is determined based upon the further utterance. One or more said unfilled slots are filled (302) with said relevant information.

Abstract: A user interface, system, and method are disclosed to facilitate specification of queries and displaying corresponding results. The user interface presents the user with dimensions that contain one or more headings arranged according to an information taxonomy, which can vary based on the intended implementation for the system and user interface. A corresponding filter or query is constructed based on the user selecting of one or more headings. The filter is applied to one or more databases to return results that satisfy the filter. The results are presented in the user interface and can include interactive items based on a particular query as well as can correspond to a fully specified task.

Abstract: According to the present invention, a method for integrating processes with a multi-faceted human centered interface is provided. The interface is facilitated to implement a hands free, voice driven environment to control processes and applications. A natural language model is used to parse voice initiated commands and data, and to route those voice initiated inputs to the required applications or processes. The use of an intelligent context based parser allows the system to intelligently determine what processes are required to complete a task which is initiated using natural language. A single window environment provides an interface which is comfortable to the user by preventing the occurrence of distracting windows from appearing. The single window has a plurality of facets which allow distinct viewing areas. Each facet has an independent process routing its outputs thereto. As other processes are activated, each facet can reshape itself to bring a new process into one of the viewing areas.

Abstract: A translation method and system include a recognition engine having a plurality of models each being employed to decode a same utterance to provide an output. A model combiner is configured to assign probabilities to each model output and configured to assign weights to the outputs of the plurality of models based on the probabilities to provide a best performing model for the context of the utterance.

Abstract: This invention relates to the field of wireless data and instant communication technologies and describes a method and a system for connecting words, phrases, or symbols of any languages or multimedia expressions, within the content of transmitted data, to telecommunication codes. The presented method of the invention selects a group of Telecom Codes, defines Content Names, assigns the Content Names to the Telecom Codes, receives the transmitted content, and redirects the content to the connected Telecom Codes after detecting the existence of the Content Names.

Abstract: The invention deals with speech recognition, such as a system for recognizing words in continuous speech. A speech recognition system is disclosed which is capable of recognizing a huge number of words, and in principle even an unlimited number of words. The speech recognition system comprises a word recognizer for deriving a best path through a word graph, and wherein words are assigned to the speech based on the best path. The word score being obtained from applying a phonemic language model to each word of the word graph. Moreover, the invention deals with an apparatus and a method for identifying words from a sound block and to computer readable code for implementing the method.

Abstract: In an embodiment, a lattice of phone strings in an input communication of a user may be recognized, wherein the lattice may represent a distribution over the phone strings. Morphemes in the input communication of the user may be detected using the recognized lattice. Task-type classification decisions may be made based on the detected morphemes in the input communication of the user.

Abstract: A method receives a request for information regarding a product or service from a user. The received request is provided to a speech processing system which attempts to generate an automated response to the received request. If the speech processing system generates a response to the received request, that response is provided to the user. However, if the speech processing system does not generate a response to the received request, the user is referred to an advisor to handle the received request.

Abstract: Methods of organizing a series of sibling data entities in a digital computer are provided for preserving sibling ranking information associated with the sibling data entities and for attaching the sibling ranking information to a joint parent of the sibling data entities to facilitate on-demand generation of ranked parent candidates. A rollup function of the present invention builds a rollup matrix (126) that embodies information about the sibling entities and the sibling ranking information and provides a method for reading out the ranked parent candidates from the rollup matrix in order of their parent confidences (141). Parent confidences are based on the sibling ranking information, either alone or in combination with n-gram dictionary ranking or other ranking information.

Abstract: Speech signal information is formatted, processed and transported in accordance with a format adapted for TCP/IP protocols used on the Internet and other communications networks. NULL characters are used for indicating the end of a voice segment. The method is useful for distributed speech recognition systems such as a client-server system, typically implemented on an intranet or over the Internet based on user queries at his/her computer, a PDA, or a workstation using a speech input interface.

Abstract: A method for processing speech audio in a network connected client device can include selecting a speech grammar for use in a speech recognition system in the network connected client device; characterizing the selected speech grammar; and, based on the characterization, determining whether to process the speech grammar locally in the network connected client device, or remotely in a speech server in the network. In one aspect of the invention, the selecting step can include establishing a communications session with a speech server; and, querying the speech server for a speech grammar over the established communications session. Additionally, the selecting step can further include registering the speech grammar in the speech recognition system.

Abstract: Embodiments are described for a voice output device having a first memory unit configured to store a plurality of audio files, a computation unit configured to associate one or more of the audio files stored in the first memory unit with one or more outputtable data records in the correct order, and an output unit having an audio data output interface for reproducing the audio files in the order prescribed by the computation unit, where the audio files comprise fixed audio files, which contain predetermined sentence components, and variable audio files, which are used to selectively supplement the fixed audio files in order to produce modularly complete sentences.

Abstract: A handheld voice activated spelling device includes a housing and a cover secured to the housing. A power source and microphone are mounted within the housing and control switches are operable from the front surface of the housing. A first memory has a plurality of words stored therein. A second memory has a plurality of word definitions stored therein, each associated with a respective word. A speech recognition apparatus is coupled to the microphone and to the first memory and responsive to the electronic signals generated by the microphone for selecting at least one word from the first memory representative of the specific word spoken by the user. A display is provided for displaying the plurality of words and the plurality of word definitions when the user operates a selected control switch. A related word classification can be also selected from a third memory and displayed on the display.

Abstract: This voice recognition method comprises a decoding stage during which an enunciated word is identified on the basis of voice signal models described with the aid of voice units, each voice signal model representing a word belonging to a predefined vocabulary, and also comprises organizing voice signal models into an optimized lexical network associated with syntactic rules during which each word is identified with a word marker, wherein temporal information is inserted within the optimized lexical network in the form of additional generic markers, so as to spot relevant moments during the decoding.

Abstract: Provided is a speech recognition device which appropriately applies limitations on target words to be recognized which are obtained from outside of the speech recognition device, as well as to eliminate the uncomfortable feeling caused by the limitation processing.