Abstract: A natural language processing system identifies an action element, a target element, and command element(s) in a text natural language command. For each identified command element, in the order it appears in the text natural language command, the natural language processing system accesses a playlist access matrix according to a matrix access counter to identify a playlist pointer associated with that command element, determines whether that playlist pointer indicates its associated command element is a best match relative to any other command elements that have already been considered and, if so, updates a playlist entry identifier with that playlist pointer and increments the matrix access counter. When all of the command elements have been considered, the natural language processing system uses the playlist entry identifier to identify a computing language command in a command playlist, and executes the computing language command on the target element based on the action element.

Abstract: The present disclosure generally relates to using voice interaction to access call functionality of a companion device. In an example process, a user utterance is received. Based on the user utterance and contextual information, the process causes a server to determine a user intent corresponding to the user utterance. The contextual information is based on a signal received from the companion device. In accordance with the user intent corresponding to an actionable intent of answering the incoming call, a command is received. Based on the command, instructions are provided to the companion device, which cause the companion device to answer the incoming call and provide audio data of the answered incoming call. Audio is outputted according to the audio data of the answered incoming call.

Abstract: A system for training a model to predict a sequence (e.g. a sequence of words) given a context is disclosed. A model can be trained to make these predictions using a combination of individual predictions compared to base truth and sequences of predictions based on previous predictions, where the resulting sequence is compared to the base truth sequence. In particular, the model can initially use the individual predictions to train the model. The model can then be further trained over the training data in multiple iterations, where each iteration includes two processes for each training element. In the first process, an initial part of the sequence is predicted, and the model and model parameters are updated after each prediction. In the second process, the entire remaining amount of the sequence is predicted and compared to the corresponding training sequence to adjust model parameters to encourage or discourage each prediction.

Abstract: Techniques for determining a command or intent likely to be subsequently invoked by a user of a system are described. A user inputs a command (either via a spoken utterance or textual input) to a system. The system determines content responsive to the command. The system also determines a second command or corresponding intent likely to be invoked by the user subsequent to the previous command. Such determination may involve analyzing pairs of intents, with each pair being associated with a probability that one intent of the pair will be invoked by a user subsequent to a second intent of the pair. The system then outputs first content responsive to the first command and second content soliciting the user as to whether the system to execute the second command.

Abstract: Provided herein is a system, method, and computer program product for modifying a language conversation model of the language learning system. Modifying the language conversation model includes receiving, using a conversational sub-system, voice inputs. The conversational sub-system converts the voice inputs to voice input data and processes the voice input data. The conversational sub-system detects an error in processing the voice input data and, based at least in part on the error, stores additional data comprising additional voice input data in a memory. The conversational sub-system applies machine learning to the additional data to derive a function that is not enabled within the language conversation model. The conversational sub-system develops an update that enables the language conversation model to implement the function. The update is applied to the language conversation model.

Abstract: A method and apparatus for managing multiple voice operation trigger phrases includes an electronic computing device enabling a first voice operation trigger phrase and a second voice operation trigger phrase on the electronic computing device. The method further includes determining whether a context applies to the electronic computing device, loading the first voice operation trigger phrase upon determining that a first context applies to the electronic computing device, and loading the second voice operation trigger phrase upon determining that a second context applies to the electronic computing device.

Abstract: Systems and methods are disclosed to implement a machine learning system that is trained to assign annotations to text fragments in an unstructured sequence of text. The system employs a neural model that includes an encoder recurrent neural network (RNN) and a decoder RNN. The input text sequence is encoded by the encoder RNN into successive encoder hidden states. The encoder hidden states are then decoded by the decoder RNN to produce a sequence of annotations for text fragments within the text sequence. In embodiments, the system employs a fixed-attention window during the decoding phase to focus on a subset of encoder hidden states to generate the annotations. In embodiments, the system employs a beam search technique to track a set of candidate annotation sequences before the annotations are outputted. By using a decoder RNN, the neural model is better equipped to capture long-range annotation dependencies in the text sequence.

Abstract: A network monitor system collects log entries from network appliances, each log entry including quantity context, first time context, first name context, and value of the quantity context. The network monitor system receives first audio signals of a spoken question and processes the audio signals to determine an ordering term and a question context. The question context includes second name context, second time context, and quantity entity context. The network monitor system compares the question context with given log entries. For each match, the network monitor system stores quantity context and value of the quantity context as a result entry in a result entries list. The network monitor system orders the result entries on the result entries list according to ordering term and pre-determined number of results and composes a response. The network monitor system converts the response to second audio signals and outputs the second audio signals.

Abstract: An information processing apparatus including: an acquisition unit configured to acquire a structure of a query sentence obtained by analyzing the input query sentence and a structure of a target document obtained by analyzing the target document obtained by search that uses a search word generated by using the query sentence; and a control unit configured to determine an answer candidate by comparing the structure of the query sentence and the structure of the target document which are acquired by the acquisition unit, and generate information for presenting a result of the determination in contrast to the query sentence.

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: Disclosed are a terminal, a server and event suggesting methods thereof, the event suggesting method using word embedding, the method including: determining user preference information about a predetermined object by assigning values to a plurality of reference items with respect to user input data corresponding to the object; generating user intent information, which includes values based on the plurality of reference items, based on at least one piece of review data corresponding to a plurality of events for recommendation objects, and accessing a database built up by adding the generated user intent information to the plurality of events; and suggesting an event following the object among the plurality of events based on a similarity between the user preference information about the object and the user intent information of each event included in the database.

Abstract: Technologies for natural language request processing include a computing device having a semantic compiler to generate a semantic model based on a corpus of sample requests. The semantic compiler may generate the semantic model by extracting contextual semantic features or processing ontologies. The computing device generates a semantic representation of a natural language request by generating a lattice of candidate alternative representations, assigning a composite weight to each candidate, and finding the best route through the lattice. The composite weight may include semantic weights, phonetic weights, and/or linguistic weights. The semantic representation identifies a user intent and slots associated with the natural language request. The computing device may perform one or more dialog interactions based on the semantic request, including generating a request for additional information or suggesting additional user intents.

Abstract: One or more computing devices, systems, and/or methods for dynamically selecting a personality for a digital assistant are provided. For example, audio associated with a conversation with a digital assistant may be received from a user. The audio may be converted into a request comprising text. A task may be determined based upon the request. One or more sentences associated with information associated with the task may be determined. A context of the conversation may be determined based upon the request and a user profile of the user. A first personality may be selected for the digital assistant from one or more personalities based upon the context of the conversation. The first personality may be used to generate audio of the one or more sentences. The audio of the one or more sentences may be presented as part of the conversation to the user.

Abstract: Non-limiting examples of the present disclosure describe generation and application of a universal hypothesis ranking model to rank/re-re-rank dialog hypotheses. An input is received through a user interface of an application for dialog processing. A plurality of dialog hypotheses are generated based on input understanding processing of the received input. The plurality of dialog hypotheses are ranked using a universal hypothesis ranking model that is applicable to a plurality of languages and locales. The ranking of the plurality of dialog hypotheses comprises using the universal hypothesis ranking model to analyze language independent features of the plurality of dialog hypotheses for policy determination. Other examples are also described including examples directed to generation of the universal hypothesis ranking model.

Abstract: An aspect provides a method, including: playing, on a display device, video content; providing, using at least one speaker, audio content associated with the video content; obtaining, from an external source, data relating to playback context; determining, using a processor, that the data relating to playback context is associated with a reduced audibility context; and providing, on the display device, textual data associated with dialog of the video content. Other aspects are described and claimed.

Abstract: Embodiments provide a non-transitory computer-readable medium containing computer program code that, when executed, performs an operation. The operation includes detecting a user action requesting an interaction with a first device and originating from a source. Additionally, embodiments determine a direction in which the source is located, relative to a current position of the first device. A response to the user action is also determined, based on a current state of the first device. Embodiments further include outputting the determined response substantially in the determined direction in which the source is located.

Abstract: According to one embodiment, an interaction apparatus includes an interaction apparatus includes a storage, a first extractor, a retriever, a generator, a second extractor and a register. The storage stores a problem and at least one solution for solving the problem. The first extractor extracts a target problem which is an expression regarded as the problem, from a first speech. The generator generates a first speech-prompting sentence prompting the user to make a speech including the target solution if the storage stores no target solution or if the user rejects the target solution. The second extractor extracts the target solution from a second speech which is a response of the user relating to the first speech-prompting sentence. The register registers, on the storage, the target problem and the target solution.

Abstract: A dialog system includes a processor. The system can further include a dialog manager. The dialog manager can be configured to receive input from a user using the processor. The system can further include a user category classification and detection module, which is configured to identify categories for the user from the received input. The system can further include a user mood detection and tracking module configured to identify a mood of the user. The system can further include a user physical and mind state and energy level detection module configured to identify a mental status of the user. The system can further include a user acquaintance module configured to identify an acquaintance status of the user. The system can further include user personality detection and tracking module configured to identify a personality status of the user. The system can further include a conversational context detection and response generation module.

Abstract: Systems, methods and computer-readable storage media for natural language understanding in combination with real-time automated humanized verification in conversation agent messaging are described.

Abstract: The present disclosure relates to remote service requesting and processing. The method includes: receiving a service processing request sent by a terminal; recognizing voice data in the service processing request to obtain voice recognition information; and sending the voice recognition information and an Internet application identifier to a corresponding third-party server according to a public identifier, so that the third-party server processes a corresponding service according to the voice recognition information. The voice data in the service processing request sent by the terminal is recognized to obtain the voice recognition information. The voice recognition information and the Internet application identifier are sent to the third-party server, so that the third-party server processes the corresponding service according to the voice recognition information.

Abstract: Systems and approaches for detecting an event, such as a crash involving a motorized and/or non-motorized vehicle, are described. In one approach, a listening component of a mobile computing device is first activated. This listening component may then monitor the sound proximate thereto (e.g., within the vehicle in which the mobile computing device is disposed). Using one or more processors, an audio source may be detected which was generated by an event (such as, for example, a crash involving the vehicle and/or an alerting triggering phrase) and was sensed by the listening component. Using an audio recognition technique, the detected audio source is processed to determine a severity of the event which generated the audio signal. This detection step may include identifying a type and magnitude of the detected audio signal using audio signal identification data. An information signal based on the severity of the event is then transmitted.

Abstract: A system for interactive voice response system crawling, comprising an IVR crawler that may be VXML, design specification, DTMF or ASR/NLSR speech-based in nature and traverses an IVR menu to discover possible interaction paths and produces test cases based on those paths, and a database that stores test cases produced by the IVR crawler during operation, and a method for using an IVR crawler to perform a system migration.

Abstract: One of the methods includes, obtaining a natural language query from a user; while converting the natural language query into structured operations to be performed on structured application programming interfaces (APIs) of a knowledge base: responsive to detecting a linguistic ambiguity associated with the converting, wherein the linguistic ambiguity identifies two or more interpretations of a word included in the natural langue query: resolving the linguistic ambiguity by selecting an interpretation of the word, from the two or more interpretations of the word, based on one or more join paths that have been generated based on a data schema of the knowledge base; modifying the converting in accordance with the selected interpretation of the word, to generate one or more structured operations; performing the one or more structured operations on the structured APIs of the knowledge base; and returning search results matching the natural language query results to the user.

Abstract: The system provided herein is a computer-implemented system that defines a social network using the linkages of technologies. According to one the teachings, the system looks for pandemic spread or integration of technology through this network to determine when an outbreak of a new technology is occurring.

Abstract: A method, apparatus, and computer program product are presented herein for user-directed reporting. In the context of a method, context information is received, wherein the context information defines the data set from which a natural language text or a graph is to be derived. The method further includes generating a set of messages based on the context information. The method further includes generating, by a processor, a context-specific report based on the set of messages and the relationship between them, wherein the context-specific report is configured to linguistically or visually express at least of messages. The method may include generating additional messages or annotations interactively. A corresponding apparatus and computer program product are provided.

Abstract: A voice command system may enable a user to select, replace, delete, and/or modify items of textual content associated with, or following, headings and/or subheadings of a document by providing an input including a command (or identifier) and a heading/subheading identifier. The system may recognize, via voice recognition software and in combination with a rule set, the command and the identifier. For example, the system may include rules for determining a particular heading/subheading of the document being referenced by the identifier. The system may also include rules for identifying the textual content related to the identified heading/subheading. Once the textual content is identified, the system may automatically select or delete the identified textual content, and/or place a cursor (for example, a text input cursor) in a location where the deleted textual content was previously located.

Abstract: A method and apparatus of a device that suggests a tokenized query completion for an input query prefix is described. In an exemplary embodiment, the device receives a query prefix from a client, wherein the query prefix includes a plurality of words. The device further generates a results set by searching a structured database using the query prefix for matches to the plurality of words in the query prefix. The device additionally determines a subset of query prefix that match specific fields of the results set by using the last N grams in the query prefix. In addition, the device ranks a tokenized query completion as a search suggestion using the query prefix, where the tokenized query completion includes a token that is a match between a matching word in the subset of query prefix and the corresponding specific field for the matching word.

Abstract: A first message may be displayed within a main chat room. The first message may be associated with a first topic. A second message may be displayed below the first message. The second message may be associated with a second topic. A third message may be received. The third message may be determined to be associated with the first topic. In response to the determining, the third message may be displayed by offsetting the third message below the first message, wherein the third message is displayed above the second message. The offsetting the third message below the first message may indicate a hierarchical relationship between the first message and the third message.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for managing dialogs. In one aspect, a method includes receiving a request associated with a task from a user device; submitting the request to each of a plurality of distinct data providers; receiving a plurality of suggested dialog responses from two or more of the data providers; scoring the one or more suggested dialog responses based on one or more scoring factors; determining a particular dialog response to provide to the user based on the scoring; and providing the determined dialog response to the user device.

Abstract: Non-limiting examples of the present disclosure describe natural language translation capabilities that enable automated process flow diagram generation from received input. Input may be received through an application for automated generation of a process flow diagram. The received input may be provided to a natural language processing component of a language understanding intelligence service. A data object, received from the natural language processing component, may be accessed. The data object provides data for creation of a process flow diagram based on the received input. In examples, the data object is generated based on natural language processing by the natural language processing component and at least one user defined grammar rule, provided by the application, for converting the received input to one or more process flow steps. The process flow diagram may be presented within the application. Other examples are also described such as reverse engineering an existing process flow diagram.

Abstract: Under one aspect, first user input including free-form text is received in a first graphical user interface (GUI). A classification engine of the computer system incorporating a machine learning model classifies words of the free-form text into a male-biased class, a female-biased class, or a neutral class. At least one of the words is classified into the male-biased class or the female-biased class. At least one of the words classified into the male-biased class or the female-biased class is flagged in the first GUI. Second user input is received in the first GUI including at least one revision to at least one of the words of the free-form text classified into the male-biased class or the female-biased class responsive to the flagging. The revised free-form text is posted to a web site for display in a second GUI.

Abstract: A premises apparatus is provided. The premises apparatus includes a processor configured to: determine that at least one monitoring service is being requested for activation and determine permit data associated with the at least one monitoring service. The processor is father configured to, if the determined permit data indicates at least one permit is required for the at least one monitoring service: transmit a request for at least one permit that is required for activation of the at least one monitoring service and, after transmitting the request for the at least one permit for activation of the at least one monitoring service, activate the at least one monitoring service.

Abstract: The technology of the present application provides a modified language model to allow for the recognition of speech containing types of disfluency. The modified language model includes a plurality of n-grams where an n-gram comprises a sequent of words. The language model also has at least one synthetic n-gram where the synthetic is a naturally occurring n-gram combined with a disfluency token. The disfluency token is representative of multiple types of disfluency and multiple pronunciations thereof.

Abstract: Disclosed herein are systems, computer-implemented methods, and computer-readable media for enhancing speech recognition accuracy. The method includes dividing a system dialog turn into segments based on timing of probable user responses, generating a weighted grammar for each segment, exclusively activating the weighted grammar generated for a current segment of the dialog turn during the current segment of the dialog turn, and recognizing user speech received during the current segment using the activated weighted grammar generated for the current segment. The method can further include assigning probability to the weighted grammar based on historical user responses and activating each weighted grammar is based on the assigned probability. Weighted grammars can be generated based on a user profile. A weighted grammar can be generated for two or more segments.

Abstract: In one embodiment, a method includes receiving, from a client system of a user of an online social network, a query inputted by the user, wherein the query comprises multiple n-grams; determining one or more head-terms and one or more modifier-terms of the n-grams based on a syntactic model; identifying one or more objects matching at least a portion of the query; ranking each identified object based on a quality of matching of the object to the determined head-terms and modifier-terms; and sending, to the client system in response to the query, a search-results interface for display, wherein the search-results interface includes one or more search results corresponding to one or more of the identified objects, respectively, each identified object corresponding to a search result having a rank greater than a threshold rank.

Abstract: Methods and systems for delivering a voice message. One method includes determining, by an electronic computing device, an environmental condition of a user. The method further includes generating, by the electronic computing device, a voice message based on the environmental condition such that the voice message includes an encoded meaning. The method further includes determining, by the electronic computing device, a voice profile of the voice message based on the environmental condition. The voice profile indicates the encoded meaning. The method further includes outputting the voice message in accordance with the voice profile with a speaker of a communication device.

Abstract: An annotation data generation assisting system includes: an input/output device receiving an input through an interactive process; morphological analysis system 380 and dependency parsing system performing morphological and dependency parsing on text data in text archive; first to fourth candidate generating units detecting a zero anaphor or a referring expression in the dependency relation of a predicate in a sequence of morphemes, identifying a position as an object of annotation and estimating candidates of expressions to be inserted by using language knowledge; a candidate DB storing estimated candidates; and an interactive annotation device reading candidates of annotation from candidate DB and annotate a candidate selected by an interactive process by input/output device.

Abstract: In a wearable translation device, the control circuit obtains an audio signal of the second language, which has been converted from a first audio signal of the first language, and obtains a second audio signal of the first language reconverted from the audio signal of the second language. The first speaker device converts the audio signal of the second language into a first voice and outputs the first voice. The second speaker device converts the second audio signal of the first language into a second voice and outputs the second voice. The audio processing circuit executes such a process that the second voice from the second speaker device is directed toward an auditory part of a user, based on a position of the auditory part relative to the second speaker device.

Abstract: An interactive response system directs input to a software-based router, which is able to intelligently respond to the input by drawing on a combination of human agents, advanced recognition and expert systems. The system utilizes human “intent analysts” for purposes of interpreting customer input. Automated recognition subsystems are trained by coupling customer input with IA-selected intent corresponding to the input, using model-updating subsystems to develop the training information for the automated recognition subsystems.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, relating to generating log-linear models. In some implementations, n-gram parameter values derived from an n-gram language model are obtained. N-gram features for a log-linear language model are determined based on the n-grams corresponding to the obtained n-gram parameter values. A weight for each of the determined n-gram features is determined, where the weight is determined based on (i) an n-gram parameter value that is derived from the n-gram language model and that corresponds to a particular n-gram, and (ii) an n-gram parameter value that is derived from the n-gram language model and that corresponds to an n-gram that is a sub-sequence within the particular n-gram. A log-linear language model having the determined n-gram features is generated, where the determined n-gram features in the log-linear language model have weights that are initialized based on the determined weights.

Abstract: A semiotic square search and/or sentiment analysis system and method are provided. In one implementation, a software implemented document search system and method are disclosed. The system and method may be used to analyze sentiments in various types of data including documents, blogs, text strings and posts.

Abstract: Systems and methods for domain classification in natural language processing based on domains are disclosed. The method includes generating a trigram corpus for the purpose of classification based on a trigram analysis of a domain model containing a hierarchical ontology and semantic construction that maps patterns of semantic tokens to syntactic patterns. An input string is parsed within each domain, tokenized in each domain. The resulting trigrams for the input text in each domain are looked up in the corresponding trigram corpus to determine the relevancy of each domain to the input text. The input string is thus classified based on the relevancy determination. The systems and methods avoids having to rely on existing annotated domain corpora for classification and allows for fast regeneration of the classifier when domain models are under frequent update and development.

Abstract: An apparatus for amending a language analysis error includes: a main language analyzer, which includes a plurality of language processing modules being sequentially connected to each other, and generates one best main analysis result for each processing module; a subsidiary language analyzer, which includes the plurality of language processing modules, and generates a plurality of subsidiary analysis results for each of the plurality of language processing modules; and an analysis result amender to in response to an error occurring in the main analysis result, acquire a subsidiary analysis result, and transmit the subsidiary analysis result to the main language analyzer.

Abstract: Methods and a system are provided for estimating automatic speech recognition (ASR) accuracy. A method includes obtaining transcriptions of utterances in a conversation over two channels. The method further includes sorting the transcriptions along a time axis using a forced alignment. The method also includes training a language model with the sorted transcriptions. The method additionally includes performing ASR for utterances in a conversation between a first user and a second user. The second user is a target of ASR accuracy estimation. The method further includes determining whether an ASR result of the second user is consistent or inconsistent with an ASR result of the first user using the trained language model. The method also includes estimating the ASR result of the second user as poor responsive to the ASR result of the second user being as inconsistent with the ASR result of the first user.

Abstract: A system and method for a highly interactive style of speech-to-speech translation is provided. The interactive procedures enable a user to recognize, and if necessary correct, errors in both speech recognition and translation, thus providing robust translation output than would otherwise be possible. The interactive techniques for monitoring and correcting word ambiguity errors during automatic translation, search, or other natural language processing tasks depend upon the correlation of Meaning Cues and their alignment with, or mapping into, the word senses of third party lexical resources, such as those of a machine translation or search lexicon. This correlation and mapping can be carried out through the creation and use of a database of Meaning Cues, i.e., SELECT. Embodiments described above permit the intelligent building and application of this database, which can be viewed as an interlingua, or language-neutral set of meaning symbols, applicable for many purposes.

Abstract: A mapping is constructed between a node of a cultural hierarchy and a node of an organizational hierarchy. The cultural hierarchy is a data structure representing an arrangement of characteristics of different social levels relative to one another in a culture. The organizational hierarchy is a data structure representing an arrangement of characteristics of different roles of people relative to one another in a group. A set of cultural attributes is constructed to represent a persona of a human user according to mapping. Using a cultural attribute, an adjustment rule is selected from a library. A machine output is modified according to the adjustment rule. The machine output is in response to a dialog between a data processing system and human user in a machine-human interaction. The adjusted output is transformed and presented as a response to the human user in the machine-human interaction.

Abstract: Co-browsing sessions provide a way for one party to see exactly what is being displayed on another party's computer or similar device. Co-browsing is particularly useful to an agent to help a customer solve a problem on a computer. Contact centers may determine portions of a co-browsing session that have been successful, capture the associated inputs, and have them play back to future customers. The agent may then be relieved from portions of the co-browsing interaction that are predictable and provide manual inputs having less predictability. The instructions for the automated co-browsing portions may be executed by the agent's device, the customer's device, and/or a server and may appear to the customer to be originating from the agent.

Abstract: Disclosed are a voice control method, device and terminal, and the method includes: after a terminal has been triggered into voice control mode, receiving input voice data and obtaining corresponding speech text according to the voice data; matching the speech text with an interface word list corresponding to a current operating interface, wherein the interface word list includes text information of the current operating interface; in response to determining that the speech text matches an interface word list successfully, obtaining in the current operating interface an operation link which corresponds to the speech text; and performing an operation corresponding to the operation link.

Abstract: Natural language query translation may be provided. A statistical model may be trained to detect domains according to a plurality of query click log data. Upon receiving a natural language query, the statistical model may be used to translate the natural language query into an action. The action may then be performed and at least one result associated with performing the action may be provided.

Abstract: A method of performing speech recognition on board an aircraft, a computer program for executing the method, as well as a speech recognition unit for performing speech recognition on board an aircraft. The method comprises receiving a speech signal spoken by a user; performing speaker recognition on the speech signal to identify the user from the speech signal; selecting a speech recognition user profile which is associated with the identified user; and performing speech recognition on the speech signal using the selected user profile.