Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech recognition using complex evolution recurrent neural networks. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A first vector sequence comprising audio features determined from the audio data is generated. A second vector sequence is generated, as output of a first recurrent neural network in response to receiving the first vector sequence as input, where the first recurrent neural network has a transition matrix that implements a cascade of linear operators comprising (i) first linear operators that are complex-valued and unitary, and (ii) one or more second linear operators that are non-unitary. An output vector sequence of a second recurrent neural network is generated. A transcription for the utterance is generated based on the output vector sequence generated by the second recurrent neural network.

Abstract: The language of a remote session can be synchronized with a client's language. A client-side synchronizer can be configured to detect when a client establishes a remote session on a server, whether to access a remote desktop or a remote application. When the client-side synchronizer detects the establishment of a remote session, it can access client-side language settings to gather language information and send it to a server-side synchronizer. The server-side synchronizer can then employ the language information to modify server-side language settings applicable the remote session so that the remote desktop or remote application will employ language settings that match the client-side language settings. In this way, the remote desktop or remote application will be in the language that the user expects even if the server's predefined language settings do not match user's language.

Abstract: Systems and methods for intelligently training a subject machine learning model includes identifying new observations comprising a plurality of distinct samples unseen by a target model during a prior training; creating an incremental training corpus based on randomly sampling a collection of training data samples that includes a plurality of new observations and a plurality of historical training data samples used in the prior training of the target model; implementing a first training mode that includes an incremental training of the target model using samples from the incremental training corpus as model training input; computing performance metrics of the target model based on the incremental training; evaluating the performance metrics of the target model against training mode thresholds; and selectively choosing based on the evaluation one of maintaining the first training mode and automatically switching to a second training mode that includes a full retraining of the target model.

Abstract: A keyboard mode switching system and method are provided. The method includes: obtaining the current window information of the currently executed window; comparing multiple records in the database according to the current window information; and when the first record corresponding to the current window information exists, switching the current keyboard operation mode currently used by the keyboard to the first keyboard operation mode in the first record so that the keyboard receives the user's input according to the first keyboard operation mode.

Abstract: In order to solve a conventional problem that, after a series of dialog between a user and a spoken dialog device has progressed to some extent, that user or another user cannot see or recognize a previous dialog status, a cross-lingual spoken dialog system is provided wherein, in a case in which an instruction from a user terminal is received by a pairing server, dialog information stored in a storage medium is transmitted to the user terminal. Accordingly, even after a series of dialog between a user and the spoken dialog device has progressed to some extent, that user or another user can see or recognize a previous dialog status.

Abstract: A method for determining whether a text complies with a query, an apparatus and a computer program product, the method comprising: obtaining a macro comprising a plurality of representations of a query; and executing the macro against a database comprising a simplified representation of at least one phrase within the text, to determine 10 whether the text complies with any of the representations of the query. The simplified representation is presented in the form of: conferrer (carrier; bearer; substrate; descriptor) and provides information pertaining to who, what, when and how with respect to the text associated with the simplified representation.

Abstract: Contextual vocabulary and methods of use are provided herein. A method may include obtaining contextual properties for a plurality of end users and device related information for computing devices used by the plurality of end users, the contextual information being obtained from a plurality of ambient information sources, including selection from a plurality of contextual repositories. Also, the method may include calculating a trust level for each of the plurality of customer analytic information sources and each of the plurality of device information databases, and building a contextual vocabulary from the contextual information and the device related information selected based upon their trust levels.

Abstract: An image forming apparatus includes one or more processors, one or more memories for storing programs including an application configured to receive a setting relating to display candidates of a plurality of languages, the programs, when executed by the one or more processors, causing the image forming apparatus to perform a process for changing a display language, and a control unit configured to control processing such that the application transmits to firmware a changing request for changing a language displayed on a screen to a different language from among the display candidates of the plurality of languages when an instruction for changing a language displayed on the screen to a different language is received.

Abstract: Intent-resolution using a phrase index may include obtaining data expressing a usage intent, the data indicating an unresolved data portion, identifying a phrase fragment based on the data expressing the usage intent and a phrase pattern, the phrase fragment including the unresolved data portion, identifying candidate tokens, identifying candidate phrases by traversing a phrase index based on the phrase fragment, wherein identifying the candidate phrases includes, in response to a determination that the phrase index includes an indexed phrase at least partially matching the phrase fragment in accordance with the phrase pattern, the indexed phrase is identified as one of the candidate phrases, identifying candidate resolved-requests, weighting and sorting the candidate tokens, the candidate phrases, and the candidate resolved-request, to obtain sorted candidate resolutions, and outputting one or more of the sorted candidate resolutions for presentation to a user as respective candidates for resolving the unres

Abstract: Systems and methods are disclosed to predict probability of occurrence of a string. A sequence of vectors is generated based at least on a maximum length of word for each symbol in the string. The sequence of vectors is provided to a machine learning unit for the string. A probability of occurrence of the string is obtained from the machine learning unit.

Abstract: A dashboard device of a vehicle includes a storage component and a processor. The storage component stores a plurality of different language packs and a first national table listing a first number of unique language indices associated respectively with the language packs. The processor is coupled to the storage component and is programmed to, in response to receipt of a machine code and a second number related to another plurality of unique language indices from a portable electronic device, use one of the language packs to set language used on the dashboard device based on content of the machine code and a relationship between the first and second numbers.

Abstract: Systems and methods for generating and publishing an app. A method in one example can include obtaining a first natural language statement to generate a first set of one or more natural language statements; translating the first natural language statement into a first version of a meaning representation; and publishing a functional version of the app based at least in part on the first version of the meaning representation so that the app is usable by one or more user devices.

Abstract: Embodiments of the systems and methods disclosed herein provide a prescriptive analytics platform and polarity analysis engine in which a user can identify a target objective and use the system to find out whether the user's objectives are being met, what predictive factors are positively or negatively affecting the targeted objectives, as well as what recommended changes the user can make to better meet the objectives. The systems and methods may include a polarity analysis engine configured to determine the polarity of terms in free-text input in view of the target objective and the predictive factors and use the polarity to generate the recommended changes.

Abstract: Additional background information is used with a trained neural network based model to help classify whether text is a subtly and/or ambiguously offensive. This additional background information can come from different sources such as the article on which the comment was made, world knowledge about the external entities (e.g., Wikipedia, Urban Dictionary), phrases referenced in the text being classified, and, the context of the previous comments/text in the thread. The background information is retrieved based on key entities (e.g., people, places things) and/or key phrases in the comment. Sentence matrix encodings are built for both the comment and the background information. The background information encoding is used to condition the comment encoding. The background information encoding, and the conditioned comment encoding are fed to a trained multi-level perceptron to classify the comment as hate speech or non-hate speech.

Abstract: A method for identifying errors associated with named entity recognition includes recognizing a candidate named entity within a text and extracting a chunk from the text containing the candidate named entity. The method further includes creating a feature vector associated with the chunk and analyzing the feature vector for an indication of an error associated with the candidate named entity. The method also includes correcting the error associated with the candidate named entity.

Abstract: Automatic semantic analysis for characterizing and correlating literary elements within a digital work of literature is accomplished by employing natural language processing and deep semantic analysis of text to create annotations for the literary elements found in a segment or in the entirety of the literature, a weight to each literary element and its associated annotations, wherein the weight indicates an importance or relevance of a literary element to at least the segment of the work of literature; correlating and matching the literary elements to each other to establish one or more interrelationships; and producing an overall weight for the correlated matches.

Abstract: A computer-implemented method, apparatus and computer program product of question generation in a cognitive system, comprising: receiving a training question; applying rules of grammar and common usage to generate variants of the training question; identifying any generalized versions of specific words, such as verbs or nouns in the training question; identifying any questions related to the training question; filtering out, utilizing a classifier, generated variants which do not make sense; and providing one or more output questions, synonymous to the training question.

Abstract: A Codex system of computers linked into a neural network continuously scans and gathers information from, understands, and interacts with, an environment, an optimizer software executing software instructions based on rules of grammar and semantics searches a encyclopedia of human knowledge to transform input into a search pattern. Subsequently the Codex monetizes and commercializes each transformed input and corresponding optimal output. A virtual maestro, artificial intelligence interaction software, uses the search pattern and optimal output in order to interact and engage scripted communication with the end user.

Abstract: A method, computer system, and a computer program product for crawling and extracting main content from a web page is provided. The present invention may include retrieving a HTML document associated with a web page. The present invention may then include identifying at least one entry point located in the retrieved HTML document by utilizing a self-adaptive entry point locator. The present invention may also include extracting a main content article associated with the retrieved HTML document based on the identified at least one entry point. The present invention may further include presenting the extracted main content associated with the retrieved HTML document to the user.

Abstract: Components of language processing engines, such as translation models and language models, can be customized for groups of users or based on user type values. Users can be organized into groups or assigned a value on a continuum based on factors such as interests, biographical characteristics, social media interactions, etc. In some implementations, translation engine components can be customized for groups of users by selecting the training data from content created by users in that group. In some implementations, the group identifier or continuum value can be part of the input into a general translation component allowing the translation component to take a language style of that user group into account when performing language processing tasks.

Abstract: Components of language processing engines, such as translation models and language models, can be customized for groups of users or based on user type values. Users can be organized into groups or assigned a value on a continuum based on factors such as interests, biographical characteristics, social media interactions, etc. In some implementations, translation engine components can be customized for groups of users by selecting the training data from content created by users in that group. In some implementations, the group identifier or continuum value can be part of the input into a general translation component allowing the translation component to take a language style of that user group into account when performing language processing tasks.

Abstract: An image processing apparatus includes a control unit that acts as: a searcher searching a first and second file, each including a text that is different in language from each other; a translator generating a translated text in the second file, by translating the text in the second file into a same language as a language of the text in the first file; a comparator comparing between the text in the first file and the translated text generated from the second file, and detecting a difference between the text in the first file and the text in the second file through comparison; and an outputter creating and outputting an image including both of the text in the first file and the text in the second file, with the difference detected by the comparator marked on the text in the first file or the text in the second file.

Abstract: Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing at least one program and a method for transforming a motion style of an avatar from a first style to a second style. The program and method include: retrieving, by a processor from a storage device, an avatar depicting motion in a first style; receiving user input selecting a second style; obtaining, based on the user input, a trained machine learning model that performs a non-linear transformation of motion from the first style to the second style; and applying the obtained trained machine learning model to the retrieved avatar to transform the avatar from depicting motion in the first style to depicting motion in the second style.

Abstract: A computer program product for computer generation of digital content, said computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processing resource to cause the processing resource to determine demographic designators associated with users of user devices that render interactive digital contents; group together similar input questions input into the interactive digital contents of the user devices; for each group of similar questions, create an answer to each group of similar questions; create a script from each answer; record each script and the corresponding group of similar input questions in a data repository; and insert a script from the data repository into a digital content template.

Abstract: Embodiments provide a user interface display method for a terminal, and a terminal. The method includes: generating, by a terminal in a first operating environment, a first user interface that includes a first input component, obtaining a first user interface picture according to the first user interface, and determining attribute information of the first input component according to a first application. The method also includes switching, by the terminal, to a second operating environment, and displaying a second user interface in the second operating environment according to the first user interface picture and the attribute information of the first input component, thereby reducing processing overheads of the terminal.

Abstract: Provided are a method for processing information and related products. The method includes the following. A target voice corresponding to a target language type is obtained. A first language type corresponding to the first headphone is determined. The target voice is converted into a first input voice corresponding to the first language type. The first input voice is sent to the first headphone.

Abstract: An information processing apparatus includes a processor that executes an information processing program to operate as a text obtaining unit that obtains a text indicating an issue occurred in an electronic apparatus, a keyword extracting unit that tokenizes the text, and filters words obtained by tokenizing the text to extract a keyword, a maintenance-information obtaining unit that obtains a maintenance-information-item or a plurality of maintenance-information-items in association with the keyword from a database that stores a plurality of maintenance-information-items, each of the plurality of maintenance-information-item being information about a solution to an issue occurred in the electronic apparatus, and a maintenance-information providing unit that provides the obtained maintenance-information-item or the plurality of obtained maintenance-information-items to a user.

Abstract: The present disclosure provides a translation system enabling a translation of a web page by an alteration of a website. The translation system comprises: a translation request receiving unit for receiving a translation request from a client device, the translation request including the URL of a web page in which text in a first language is displayed; a translating unit for translating the text in the first language included in the web page indicated by the URL into text in a second language by referring to a bilingual database storing words and phrases in the first language associated with words and phrases in the second language constituting translated words and phrases of the words and phrases in the first language; and a translation sending unit for sending the translated text in the second language to the client device.

Abstract: One or more lines of computer code are accessed. An electronic dictionary file is retrieved in response to the accessing of the computer code. The electronic dictionary file contains definitions for a plurality of commands or terms associated with the one or more lines of computer code. Based on the definitions contained in the electronic dictionary file, the one or more lines of computer code are parsed. An output is generated based on the parsing of the computer code. The output contains information explaining the one or more lines of computer code or an intended result of an execution thereof.

Abstract: A server receives a web-protocol request from a client that triggers a server-side operation in response to the request. Based on a result of the operation, the server identifies first text information to be returned to the client. A language associated with the client device is determined. A determination is made as to whether a resource file corresponding to the determined language is loaded in a volatile memory on the server. In a case where a resource file corresponding to the determined language is not loaded in the volatile memory, the resource file corresponding to the determined language is loaded from a non-volatile memory into the volatile memory on the server. Second text information which corresponds to the first text information to be returned to the client device is retrieved from the resource file in the volatile memory. The second text information is transmitted to the client.

Abstract: Methods, apparatuses, and computer program product are described herein configured to transform an input model into a format that can be expressed causally in one or more phrases with a narrative representation. The methods, apparatuses, and computer program product may be configured to obtain the input model representing a data set composed of a plurality of nodes, a plurality of edges, and a plurality of paths and identify a path of maximal significance. The methods, apparatuses, and computer program product include identifying a subset of nodes from the plurality of nodes and a subset of edges from the plurality of edges associated with the path of maximal significance. The methods, apparatuses, and computer program product are configured to generate, using a natural language generation system, one or more phrases with a narrative representation describing the subset of nodes and the subset of edges associated with the path of maximal significance.

Abstract: A configuration is implemented to receive, with a processor from a customer care platform, a request for spoken language interpretation of a user query from a first spoken language to a second spoken language. The first spoken language is spoken by a user situated at a display-based device that is remotely situated from the customer care platform. The user query is sent from the display-based device by the user to the customer care platform. The configuration performs, at a language interpretation platform, a first spoken language interpretation of the user query from the first spoken language to the second spoken language. Further, the configuration transmits, from the language interpretation platform to the customer care platform, the first spoken language interpretation so that a customer care representative speaking the second spoken language understands the first spoken language being spoken by the user.

Abstract: A method for providing an adaptive keyboard interface is provided, which includes searching for at least one candidate message that is expected to be prepared by a user with reference to context information about a situation in which the user prepares a message; and if a first text object of text objects that are included in the at least one searched candidate message is selected by the user, providing at least one replacement text object having a degree of association with the first text object, which is equal to or higher than a predetermined level, based on the first text object.

Abstract: Systems and methods are described for customizing a display of media asset information based on a user preference profile. In particular, media asset description information is customized by changing a term to a synonym that matches an event or keyword in the user preference profile.

Abstract: A display apparatus is provided. The display apparatus according to an embodiment includes a display, and a processor configured to control the display to display a UI screen including a plurality of text objects, control the display to display a text object in a different language from a preset language among the plurality of text objects, along with a preset number, and in response to a recognition result of a voice uttered by a user including the displayed number, perform an operation relating to a text object corresponding to the displayed number.

Abstract: Systems, methods, and devices are configured to manage the language used for documentation in a multi-function device (MFD). They include providing a message to a user, the message comprising a selection to choose a language as part of the initialization of the MFD. They also include receiving an input including the language chosen by the user. They further include determining if the language chosen by the user matches a default language in the MFD and modifying code associated with settings in electronic documentation to enable use of the language chosen by the user during the initialization in place of the default language in the electronic documentation when it is determined that the language chosen by the user does not match the default language, the electronic documentation providing information to the user for operation of the MFD.

Abstract: Implementations are directed to receiving text data including a string of characters, processing the text data to determine a set of reference scores including two or more reference scores, each reference score being associated with a respective language, and being determined based on the text data and a dictionary document provided for the respective language, each dictionary document including a compression of a language document provided in the respective language, selectively determining a language of the text data based on the set of reference scores, and providing language data representative of the language as output.

Abstract: Methods, systems, and devices for managing robot resources are described. A robot receives from an application a request to reserve a particular set of physical resources of the robot. The robot then determines that each of the physical resources in the set are available to the application and, based on the determination, allocates exclusive use of the particular set of resources to the application by (i) generating a token corresponding to the set of resources, (ii) providing the token to the application, and (iii) updating token data that associates the token with the set of resources. The robot then controls access to the particular set of resources such that, while token data indicates that the token is valid, commands from applications that involve the set of resources are only executed when provided with the token corresponding to the allocation of access to the particular set of resources.

Abstract: A multi-mode voice controlled user interface is described. The user interface is adapted to conduct a speech dialog with one or more possible speakers and includes a broad listening mode which accepts speech inputs from the possible speakers without spatial filtering, and a selective listening mode which limits speech inputs to a specific speaker using spatial filtering. The user interface switches listening modes in response to one or more switching cues.

Abstract: A computer-implemented method includes: obtaining and storing, by a computer device, a service profile defining writing style rules, scoring information, and threshold information; receiving, by the computer device, an extracted program integrated information (PII) message from a client device; determining, by the computer device, an overall validation score of the extracted PII message is less than a threshold defined by the threshold information; generating, by the computer device, a modified PII message by changing content of the extracted PII message; and returning, by the computer device, the modified PII message to the client device for replacing the extracted PII message in a source code.

Abstract: Aspects include learning object/action pairs for recipe ingredients. A method includes extracting an ingredient list that includes a plurality of entries from a recipe stored in an electronic format. Each of the plurality of entries includes one or more words. For each multi-word entry in the ingredient list, an ingredient associated with the entry is identified by comparing words contained in the entry to terms in a lexicon of cooking domain terms. It is determined whether an action corresponds to the identified ingredient. The determining includes parsing the entry to locate a verb that is associated with the identified ingredient. Based on determining that an action corresponds to the identified ingredient, a food item action pair that includes the identified ingredient and the located verb is output.

Abstract: Embodiments for semantic hierarchical grouping of short text fragments by a processor. Sub-terms are extracted from a plurality of input text fragments according to a lexical sub-term hierarchy. Each of the sub-terms in the lexical sub-term hierarchy are matched with concepts based on an ontology of concepts representing a domain knowledge. The input text fragments are automatically grouped into a hierarchy of concepts based on the matching and a semantical relationship between each concept and matching sub-term.

Abstract: A method for operating a transaction processing system to provide location based language capabilities and dynamic receipts in connection with money transfer services. One embodiment includes storing, in a database, information representative of transaction location based attributes and associated one or more languages, and storing information representative of receipt field attributes and associated receipt fields. Send and/or receive transaction requests from points of sale are received at a transaction processor. The received transaction requests are processed using the stored information in the database to identify transaction location attributes of the requests and the associated one or more languages, and processed using the information stored in the database to identify receipt field attributes of the requests and to determine the associated receipt fields.

Abstract: Systems and methods for editing data files. One system includes a processor. The processor is configured to receive reviewer feedback associated with a first portion of content included in a data file, analyze the reviewer feedback using natural language processing to determine an edit to the first portion of the content included in the data file, and apply the edit to the first portion of the content included in the data file. The processor is also configured to determine a second portion of the content included in the data file based on the reviewer feedback and apply the edit to the second portion of the content included in the data file.

Abstract: The present invention allows text analysis and routing of an outgoing message. The system intercepts outgoing messages for analysis by a TAS software module. The module assigns an analytical score to the message, then compares the score to a threshold. If the score is below the threshold, the message is transmitted to its ultimate destination. If not, the message may be routed for correction by the message's composer or quality assurance staff. After such correction, the message new analytical score is generated and compared, and, if necessary, the process repeats again.

Abstract: A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process including detecting, from a text, one of a plurality of first strings included in first information, each of the plurality of first strings being associated with control information, determining whether to execute a determination process based on the control information associated with a specified string when the specified string is detected from the text, the determination process determining whether the second string is included in second information, the second information including at least one of third string being associated with another coded string, the second string being a string that includes the specified string at the beginning of the second string, executing the determination process when the determination process is determined to be executed and transforming the second string into the another coded string associated with the third string.

Abstract: A virtual vocabulary database is provided for use with a with a particular user database as part of a speech recognition system. Vocabulary elements within the virtual database are imported from the user database and are tagged to include numerical data corresponding to the historical use of the vocabulary element within the user database. For each speech input, potential vocabulary element matches from the speech recognition system are provided to the virtual database software which creates virtual sub-vocabularies from the criteria according to predefined criteria templates. The software then applies vocabulary element weighting adjustments according to the virtual sub-vocabulary weightings and applies the adjustment to the default weighting provided by the speech recognition system. The modified weightings are returned with the associated vocabulary elements to the speech engine for selection of an appropriate match to the input speech.

Abstract: A language learning system is provided. The language learning system includes a computing device and a language learning file. The computing device includes a general user interface and a central processing unit. The language learning file is run by the central processing unit and includes multilingual data, a language display module, and a language learning module. The multilingual data includes a graphical constant element and dynamic lexical elements corresponding to the graphical constant element and having native language data and secondary language data translating from the native language data. The display module displays the graphical constant element and a native language element of the native language data through the general user interface in juxtaposed position. The language learning module switches between displays of the native language element with a secondary language element of the secondary language data.

Abstract: A computer system configured to integrate a plurality of websites is provided. The computer system includes a memory, a display, and at least one processor coupled to the memory and the display. In one example, the system comprises a monitoring component executed by the at least one processor and configured to analyze web content generated by a website provider and detect one or more identified elements within the web content. The system may also comprise a converting component executed by the at least one processor and configured to, responsive to detection of the one or more identified elements by the monitoring component, convert at least one portion of the web content into converted content, wherein the at least one processor is configured to display, in the display, the converted content and at least one other portion of the web content.

Abstract: Artificial intelligence (AI) technology can be used in combination with composable communication goal statements and an ontology to facilitate a user's ability to quickly structure story outlines in a manner usable by an NLG narrative generation system without any need for the user to directly author computer code. This AI technology permits NLG systems to determine the appropriate content for inclusion in a narrative story about a data set in a manner that will satisfy a desired communication goal. The AI technology also permits story edits to cause corresponding updates to an ontology and/or story outline used to support narrative generation.