Abstract: Data associated with spoken language may be obtained. An analysis of the obtained data may be initiated for understanding of the spoken language using a deep convex network that is integrated with a kernel trick. The resulting kernel deep convex network may also be constructed by stacking one shallow kernel network over another with concatenation of the output vector of the lower network with the input data vector. A probability associated with a slot that is associated with slot-filling may be determined, based on local, discriminative features that are extracted using the kernel deep convex network.

Abstract: A method for dialog agent integration comprises discovering a dialog agent required for a dialog request including dialog information comprising terms required for audio feedback in a service domain required for the dialog request, extracting the dialog information from the discovered dialog agent, integrating the dialog information to existing dialog information of a dialog system (DS) that provides dialog functionality for an electronic device, and expanding the service domain dialog functionality of the DS with the integrated dialog information.

Abstract: Systems and methods to provide a set of dictionaries and highlighting lists for speech recognition and highlighting, where the speech recognition focuses only on a limited scope of vocabulary as present in a document. The systems and methods allow a rapid and accurate matching of the utterance with the available text, and appropriately indicate the location in the text or signal any errors made during reading. Described herein is a system and method to create speech recognition systems focused on reading a fixed text and providing feedback on what they read to improve literacy, aid those with disabilities, and to make the reading experience more efficient and fun.

Abstract: An image processing apparatus including: image processor which processes broadcasting signal, to display image based on processed broadcasting signal; communication unit which is connected to a server; a voice input unit which receives a user's speech; a voice processor which processes a performance of a preset corresponding operation according to a voice command corresponding to the speech; and a controller which processes the voice command corresponding to the speech through one of the voice processor and the server if the speech is input through the voice input unit. If the voice command includes a keyword relating to a call sign of a broadcasting channel, the controller controls one of the voice processor and the server to select a recommended call sign corresponding to the keyword according to a predetermined selection condition, and performs a corresponding operation under the voice command with respect to the broadcasting channel of the recommended call sign.

Abstract: Techniques disclosed herein include systems and methods for creating a bootstrap call router for other languages by using selected N-best translations. Techniques include using N-best translations from a machine translation system so as to increase a possibility that desired keywords in a target language are covered in the machine translation output. A 1-best translation is added to a new text corpus. This is followed by selecting a subset that provides a varied set of translations for a given source transcribed utterance for better translation coverage. Additional translations are added to the new text corpus based on a measure of possible translations having words not yet seen for the selected transcribed utterances, and also based on possible translation having words that are not associated with very many or semantic tags in the new text corpus. Candidate translations can be selected from a window of N-best translations calculated based on machine translation accuracy.

Abstract: A language model learning system for learning a language model on an identifiable basis relating to a word error rate used in speech recognition. The language model learning system (10) includes a recognizing device (101) for recognizing an input speech by using a sound model and a language model and outputting the recognized word sequence as the recognition result, a reliability degree computing device (103) for computing the degree of reliability of the word sequence, and a language model parameter updating device (104) for updating the parameters of the language model by using the degree of reliability. The language model parameter updating device updates the parameters of the language model to heighten the degree of reliability of the word sequence the computed degree of reliability of which is low when the recognizing device recognizes by using the updated language model and the reliability degree computing device computes the degree of reliability.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing speech recognition using models that are based on where, within a building, a speaker makes an utterance are disclosed. The methods, systems, and apparatus include actions of receiving data corresponding to an utterance, and obtaining location indicia for an area within a building where the utterance was spoken. Further actions include selecting one or more models for speech recognition based on the location indicia, wherein each of the selected one or more models is associated with a weight based on the location indicia. Additionally, the actions include generating a composite model using the selected one or more models and the respective weights of the selected one or more models. And the actions also include generating a transcription of the utterance using the composite model.

Abstract: Embodiments of the present invention provide a method, system and computer program product for the automated voice enablement of a Web page. In an embodiment of the invention, a method for voice enabling a Web page can include selecting an input field of a Web page for speech input, generating a speech grammar for the input field based upon terms in a core attribute of the input field, receiving speech input for the input field, posting the received speech input and the grammar to an automatic speech recognition (ASR) engine and inserting a textual equivalent to the speech input provided by the ASR engine into a document object model (DOM) for the Web page.

Abstract: Disclosed herein are systems, methods, and computer readable-media for providing an automatic synthetically generated voice describing media content, the method comprising receiving one or more pieces of metadata for a primary media content, selecting at least one piece of metadata for output, and outputting the at least one piece of metadata as synthetically generated speech with the primary media content. Other aspects of the invention involve alternative output, output speech simultaneously with the primary media content, output speech during gaps in the primary media content, translate metadata in foreign language, tailor voice, accent, and language to match the metadata and/or primary media content. A user may control output via a user interface or output may be customized based on preferences in a user profile.

Abstract: Systems and methods for receiving natural language queries and/or commands and execute the queries and/or commands. The systems and methods overcomes the deficiencies of prior art speech query and response systems through the application of a complete speech-based information query, retrieval, presentation and command environment. This environment makes significant use of context, prior information, domain knowledge, and user specific profile data to achieve a natural environment for one or more users making queries or commands in multiple domains. Through this integrated approach, a complete speech-based natural language query and response environment can be created. The systems and methods creates, stores and uses extensive personal profile information for each user, thereby improving the reliability of determining the context and presenting the expected results for a particular question or command.

Abstract: A system and method for processing multi-modal device interactions in a natural language voice services environment may be provided. In particular, one or more multi-modal device interactions may be received in a natural language voice services environment that includes one or more electronic devices. The multi-modal device interactions may include a non-voice interaction with at least one of the electronic devices or an application associated therewith, and may further include a natural language utterance relating to the non-voice interaction. Context relating to the non-voice interaction and the natural language utterance may be extracted and combined to determine an intent of the multi-modal device interaction, and a request may then be routed to one or more of the electronic devices based on the determined intent of the multi-modal device interaction.

Abstract: A method includes receiving script data including script words for dialogue, receiving audio data corresponding to at least a portion of the dialogue, wherein the audio data includes timecodes associated with dialogue words, generating a sequential alignment of the script words to the dialogue words, matching at least some of the script words to corresponding dialogue words to determine hard alignment points, partitioning the sequential alignment of script words into alignment sub-sets, wherein the bounds of the alignment sub-subsets are defined by adjacent hard-alignment points, and wherein the alignment subsets includes a sub-set of the script words and a corresponding sub-set of dialogue words that occur between the hard-alignment points, determining corresponding timecodes for a sub-set of script words in a sub-subset based on the timecodes associated with the sub-set of dialogue words, and generating time-aligned script data including the sub-set of script words and their corresponding timecodes.

Abstract: In a particular embodiment, a method includes receiving a communication from a source. The communication is associated with an account of the source. The method further includes determining, at a processor, whether a time period satisfies a threshold time period. The time period is associated with a voice message box having a particular status. The method further includes routing the communication to a first destination based on a determination that the time period satisfies the threshold time period.

Abstract: Audio content is converted to text using speech recognition software. The text is then associated with a distinct voice or a generic placeholder label if no distinction can be made. From the text and voice information, a word cloud is generated based on key words and key speakers. A visualization of the cloud displays as it is being created. Words grow in size in relation to their dominance. When it is determined that the predominant words or speakers have changed, the word cloud is complete. That word cloud continues to be displayed statically and a new word cloud display begins based upon a new set of predominant words or a new predominant speaker or set of speakers. This process may continue until the meeting is concluded. At the end of the meeting, the completed visualization may be saved to a storage device, sent to selected individuals, removed, or any combination of the preceding.

Abstract: Embodiments are directed towards an automated machine learning framework to extract keywords within a message that are relevant to an attachment to the message. The machine learning model finds a set of relevant sentences within the message determined to be relevant to the one or more attachments based on identification of one or more sentence level features within a given sentence. The sentence level features include, for example, anchor features, noisy sentence features, short message features, threading features, anaphora detections, and lexicon features. From the set of relevant sentences, useful keywords may be extracted using a sequence of heuristics to convert the sentence set into the set of useful keywords. The set of useful keywords may then be associated to at least one attachment such that the keywords may subsequently be used to perform various indexing, searching, sorting, and to provide further context to the attachment.

Abstract: Techniques for conversion of non-back-off language models for use in speech decoders. For example, an apparatus for conversion of non-back-off language models for use in speech decoders. For example, an apparatus is configured convert a non-back-off language model to a back-off language model. The converted back-off language model is pruned. The converted back-off language model is usable for decoding speech.

Abstract: A method for real-time data-pattern analysis. The method includes receiving and queuing at least one data-pattern analysis request by a data-pattern analysis unit controller. At least one data stream portion is also received and stored by the data-pattern analysis unit controller, each data stream portion corresponding to a received data-pattern analysis request. Next, a received data-pattern analysis request is selected by the data-pattern analysis unit controller along with a corresponding data stream portion. A data-pattern analysis is performed based on the selected data-pattern analysis request and the corresponding data stream portion, wherein the data-pattern analysis is performed by one of a plurality of data-pattern analysis units.

Abstract: A plurality of statements are received from within a grammar structure. Each of the statements is formed by a number of word sets. A number of alignment regions across the statements are identified by aligning the statements on a word set basis. Each aligned word set represents an alignment region. A number of potential confusion zones are identified across the statements. Each potential confusion zone is defined by words from two or more of the statements at corresponding positions outside the alignment regions. For each of the identified potential confusion zones, phonetic pronunciations of the words within the potential confusion zone are analyzed to determine a measure of confusion probability between the words when audibly processed by a speech recognition system during the computing event. An identity of the potential confusion zones across the statements and their corresponding measure of confusion probability are reported to facilitate grammar structure improvement.

Abstract: Disclosed herein are systems, computer-implemented methods, and computer-readable storage media for recognizing speech by adapting automatic speech recognition pronunciation by acoustic model restructuring. The method identifies an acoustic model and a matching pronouncing dictionary trained on typical native speech in a target dialect. The method collects speech from a new speaker resulting in collected speech and transcribes the collected speech to generate a lattice of plausible phonemes. Then the method creates a custom speech model for representing each phoneme used in the pronouncing dictionary by a weighted sum of acoustic models for all the plausible phonemes, wherein the pronouncing dictionary does not change, but the model of the acoustic space for each phoneme in the dictionary becomes a weighted sum of the acoustic models of phonemes of the typical native speech. Finally the method includes recognizing via a processor additional speech from the target speaker using the custom speech model.

Abstract: Systems, methods, and computer program products for machine translation are provided. In some implementations a system is provided. The system includes a language model including a collection of n-grams from a corpus, each n-gram having a corresponding relative frequency in the corpus and an order n corresponding to a number of tokens in the n-gram, each n-gram corresponding to a backoff n-gram having an order of n?1 and a collection of backoff scores, each backoff score associated with an n-gram, the backoff score determined as a function of a backoff factor and a relative frequency of a corresponding backoff n-gram in the corpus.

Abstract: Determining synonyms of words in a set of documents. Particularly, when provided with a word or phrase as input, in exemplary embodiments there is afforded the return of a predetermined number of “top” synonym words (or phrases) for an input word (or phrase) in a specific collection of text documents. Further, a user is able to provide ongoing and iterative positive or negative feedback on the returned synonym words, by manually accepting or rejecting such words as the process is underway.

Abstract: A method for executing a fully mixed initiative dialogue (FMID) interaction between a human and a machine, a dialogue system for a FMID interaction between a human and a machine and a computer readable data storage medium having stored thereon computer code for instructing a computer processor to execute a method for executing a FMID interaction between a human and a machine are provided. The method includes retrieving a predefined grammar setting out parameters for the interaction; receiving a voice input; analyzing the grammar to dynamically derive one or more semantic combinations based on the parameters; obtaining semantic content by performing voice recognition on the voice input; and assigning the semantic content as fulfilling the one or more semantic combinations.

Abstract: In a method of creating a natural language (NL) macro, a first term/phrase and a second term/phrase in an imprecise syntax are obtained, and an association between the first and the second terms/phrases is created. The association is stored as an NL macro. In a method of using an NL macro in an NL query, it is determined that an original NL query includes an NL macro, and the NL macro is replaced with its corresponding NL value to form a revised NL query. The revised NL query is processed to generate one or more answers.

Abstract: A methodology to improve a listeners understanding of the spoken word. The methodology comprises a large set of vocal recordings organized by categories or topics. The methodology allows the user to select a type of recording to be played and then allows the users to test the listener's ability to understand the audio recording through a user input and verification process. The methodology may also allow for the user to listen to the recording multiple times and an insert via typed words what the listeners thinks they heard. The methodology may also allow the listeners to submit the typed words and receive an evaluation as to the listener's ability to correctly hear the spoken word.

Abstract: A voice talk function-enabled terminal and voice talk control method for outputting distinct content based on the current emotional state, age, and gender of the user are provided. The mobile terminal supporting a voice talk function includes a display unit, an audio processing unit, which selects content corresponding to a first criterion associated with a user in response to a user input, determines a content output scheme based on a second criterion associated with the user, and outputs the selected content through the display unit and audio processing unit according to the content output scheme.

Abstract: Disclosed herein are systems and methods to incorporate human knowledge when developing and using statistical models for natural language understanding. The disclosed systems and methods embrace a data-driven approach to natural language understanding which progresses seamlessly along the continuum of availability of annotated collected data, from when there is no available annotated collected data to when there is any amount of annotated collected data.

Abstract: Topics of potential interest to a user, useful for purposes such as targeted advertising and product recommendations, can be extracted from voice content produced by a user. A computing device can capture voice content, such as when a user speaks into or near the device. One or more sniffer algorithms or processes can attempt to identify trigger words in the voice content, which can indicate a level of interest of the user. For each identified potential trigger word, the device can capture adjacent audio that can be analyzed, on the device or remotely, to attempt to determine one or more keywords associated with that trigger word. The identified keywords can be stored and/or transmitted to an appropriate location accessible to entities such as advertisers or content providers who can use the keywords to attempt to select or customize content that is likely relevant to the user.

Abstract: In an embodiment, a method comprises analyzing a string of text that was generated based on audio input, identifying a plurality of text segments, wherein each text segment of the plurality of text segments comprises one or more words in the string of text, wherein at least one of the plurality of segments comprises a plurality of words, and organizing the plurality of text segments into a list of items, wherein each segment is a separate item in the list.

Abstract: A method and a system for a speech recognition system, comprising an electronic speech-based document is associated with a document template and comprises one or more sections of text recognized or transcribed from sections of speech. The sections of speech are transcribed by the speech recognition system into corresponding sections of text of the electronic speech based document. The method includes the steps of dynamically creating sub contexts and associating the sub context to sections of text of the document template.

Abstract: Systems and methods for improving communication over a network are provided. A system for improving communication over a network, comprises a detection module capable of detecting data indicating a problem with a communication between at least two participants communicating via communication devices over the network, a management module capable of analyzing the data to determine whether a participant is dissatisfied with the communication, wherein the management module includes a determining module capable of determining that the participant is dissatisfied, and identifying an event causing the dissatisfaction, and a resolution module capable of providing a solution for eliminating the problem.

Abstract: Disclosed are techniques and systems to provide a narration of a text in multiple different voices. In some aspects, systems and methods described herein can include receiving a user-based selection of a first portion of words in a document where the document has a pre-associated first voice model and overwriting the association of the first voice model, by the one or more computers, with a second voice model for the first portion of words.

Abstract: A display device with a voice recognition capability may be used to allow a user to speak voice commands for controlling certain features of the display device. As a means for increasing operational efficiency, the display device may utilize a plurality of voice recognition units where each voice recognition unit may be assigned a specific task.

Abstract: A display device with a voice recognition capability may be used to allow a user to speak voice commands for controlling certain features of the display device. As a means for increasing operational efficiency, the display device may utilize a plurality of voice recognition units where each voice recognition unit may be assigned a specific task.

Abstract: A method for processing speech, comprising semantically parsing a received natural language speech input with respect to a plurality of predetermined command grammars in an automated speech processing system; determining if the parsed speech input unambiguously corresponds to a command and is sufficiently complete for reliable processing, then processing the command; if the speech input ambiguously corresponds to a single command or is not sufficiently complete for reliable processing, then prompting a user for further speech input to reduce ambiguity or increase completeness, in dependence on a relationship of previously received speech input and at least one command grammar of the plurality of predetermined command grammars, reparsing the further speech input in conjunction with previously parsed speech input, and iterating as necessary. The system also monitors abort, fail or cancel conditions in the speech input.

Abstract: In addition to conveying primary information, human speech also conveys information concerning the speaker's gender, age, socioeconomic status, accent, language spoken, emotional state, or other personal characteristics, which is referred to as secondary information. Disclosed herein are both the means of automatic discovery and use of such secondary information to direct other aspects of the behavior of a controlled system. One embodiment of the invention comprises an improved method to determine, with high reliability, the gender of an adult speaker. A further embodiment of the invention comprises the use of this information to display a gender-appropriate advertisement to the user of an information retrieval system that uses a cell phone as the input and output device.

Abstract: Disclosed herein are systems, computer-implemented methods, and tangible computer-readable storage media for generating a natural language spoken dialog system. The method includes nominating a set of allowed dialog actions and a set of contextual features at each turn in a dialog, and selecting an optimal action from the set of nominated allowed dialog actions using a machine learning algorithm. The method includes generating a response based on the selected optimal action at each turn in the dialog. The set of manually nominated allowed dialog actions can incorporate a set of business rules. Prompt wordings in the generated natural language spoken dialog system can be tailored to a current context while following the set of business rules. A compression label can represent at least one of the manually nominated allowed dialog actions.

Abstract: Computer implemented speech processing generates one or more pronunciations of an input word in a first language by a non-native speaker of the first language who is a native speaker of a second language. The input word is converted into one or more pronunciations. Each pronunciation includes one or more phonemes selected from a set of phonemes associated with the second language. Each pronunciation is associated with the input word in an entry in a computer database. Each pronunciation in the database is associated with information identifying a pronunciation language and/or a phoneme language.

Abstract: An apparatus for providing a control input signal for an industrial process or technical system having one or more controllable elements includes elements for generating a semantic space for a text corpus, and elements for generating a norm from one or more reference words or texts, the or each reference word or text being associated with a defined respective value on a scale, and the norm being calculated as a reference point or set of reference points in the semantic space for the or each reference word or text with its associated respective scale value. Elements for reading at least one target word included in the text corpus, elements for predicting a value of a variable associated with the target word based on the semantic space and the norm, and elements for providing the predicted value in a control input signal to the industrial process or technical system.

Abstract: Systems and methods for filtering media containing objectionable content are described. Marker files that list the times objectionable content occurs in audio content (such as a song, podcast, audio associated with a video or television program, or the like) can be stored in a user device. When a user plays audio content for which a marker file exists, the system can automatically filter out the objectionable content marked in the marker file from playback of the audio content. The system may also provide functionality for the user to specify a level of filtering to be applied or even specific words to be filtered from audio content.

Abstract: Identify, Capture, Retain and Synthesize Non-Linguistic and Discourse Components of Speech across Languages

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: A facility for conducting a real-time conversation in which the selected one of a number of participants utilizes a silent mode is described. Remark spoken by participants other than the selected one are transformed into text and displayed for the selected participant. Remarks entered textually by the selected participant are transformed into speech and played audibly for participants other than the selected one.

Abstract: It is possible to provide meaningful responses to queries using systems which consider usage of words in the queries when analyzing those queries and determining what information is possibly relevant. This approach can be applied in online shopping systems by identification of nouns or noun phrases reflecting products available through the system.

Abstract: At least one computer-mediated communication produced by or received by an author is collected and parsed to identify categories of information within it. The categories of information are processed with at least one analysis to quantify at least one type of information in each category. A first output communication is generated regarding the at least one computer-mediated communication, describing the psychological state, attitudes or characteristics of the author of the communication. A second output communication is generated when a difference between the quantification of at least one type of information for at least one category and a reference for the at least one category is detected involving a psychological state, attitude or characteristic of the author to which a responsive action should be taken.

Abstract: It comprises analyzing audio content of multimedia files and performing a speech to text transcription thereof automatically by means of an ASR process, and selecting acoustic and language models adapted for the ASR process at least before the latter processes the multimedia file, i.e. “a priori”. The method is particularly applicable to the automatic indexing, aggregation and clustering of news from different sources and from different types of files, including text, audio and audiovisual documents without any manual annotation.

Abstract: A system, method and computer readable medium that provides an automated web transcription service is disclosed. The method may include receiving input speech from a user using a communications network, recognizing the received input speech, understanding the recognized speech, transcribing the understood speech to text, storing the transcribed text in a database, receiving a request via a web page to display the transcribed text, retrieving transcribed text from the database, and displaying the transcribed text to the requester using the web page.

Abstract: A natural language dialogue system and a method capable of correcting a speech response are provided. The method includes following steps. A first speech input is received. At least one keyword included in the first speech input is parsed to obtain a candidate list having at least one report answers. One of the report answers is selected from the candidate list as a first report answer, and a first speech response is output according to the first report answer. A second speech input is received and parsed to determine whether the first report answer is correct. If the first report answer is incorrect, another report answer other than the first report answer is selected from the candidate list as a second report answer. According to the second report answer, a second speech response is output.

Abstract: A natural language dialogue method and a natural language dialogue system are provided. In the method, a first speech input is received and parsed to generate at least one keyword included in the first speech input, so that a candidate list including at least one report answer is obtained. According to a properties database, one report answer is selected from the candidate list, and a first speech response is output according to the report answer. Other speech inputs are received, and a user's preference data is captured from the speech inputs. The user's preference data is stored in the properties database.

Abstract: Facts are extracted from speech and recorded in a document using codings. Each coding represents an extracted fact and includes a code and a datum. The code may represent a type of the extracted fact and the datum may represent a value of the extracted fact. The datum in a coding is rendered based on a specified feature of the coding. For example, the datum may be rendered as boldface text to indicate that the coding has been designated as an “allergy.” In this way, the specified feature of the coding (e.g., “allergy”-ness) is used to modify the manner in which the datum is rendered. A user inspects the rendering and provides, based on the rendering, an indication of whether the coding was accurately designated as having the specified feature. A record of the user's indication may be stored, such as within the coding itself.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating direct speech messages based on voice commands that include indirect speech messages. In one aspect, a method includes receiving a voice input corresponding to an utterance. A determination is made whether a transcription of the utterance includes a command to initiate a communication to a user and a segment that is classified as indirect speech. In response to determining that the transcription of the utterance includes the command and the segment that is classified as indirect speech, the segment that is classified as indirect speech is provided as input to a machine translator. In response to providing the segment that is classified as indirect speech to the machine translator, a direct speech segment is received from the machine translator. A communication is initiated that includes the direct speech segment.