Abstract: In some implementations, a language proficiency of a user of a client device is determined by one or more computers. The one or more computers then determines a text segment for output by a text-to-speech module based on the determined language proficiency of the user. After determining the text segment for output, the one or more computers generates audio data including a synthesized utterance of the text segment. The audio data including the synthesized utterance of the text segment is then provided to the client device for output.

Abstract: Systems, devices and methods are described herein for segmentation of content, and more specifically for segmentation of content in a content management system. In one aspect, a method may include receiving content associated with speech, text, or closed captioning data. The speech, the text, or the closed captioning data may be analyzed to derive at least one of a topic, subject, or event for at least a portion of the content. The content may be divided into two or more content segments based on the analyzing. At least one of the topic, the subject, or the event may be associated with at least one of the two or more content segments based on the analyzing. At least one of the two or more content segments may then be published such that each of the two or more content segments is individually accessible.

Abstract: One embodiment provides a method, including: collecting, at an information handling device, at least one signal received from a living object, in response to an event associated with the living object, wherein the living object is in a communicative state with a person talking to the living object; extracting, using a processor, a set of predetermined features from the signal collected; and determining, responsive to extracting, an intent associated with the living object in response to the event posed to the living object from the strength of the signal. Other embodiments are disclosed and described.

Abstract: Techniques for automated training content generation are provided. A plurality of questions are retrieved, where each of the plurality of questions is associated with an answer in a plurality of answers. Further, it is determined that a first and a second answer in the plurality of answers are equivalent. A first question corresponding to the first answer and a second question corresponding to the second answer are identified, and a first question cluster including the first question and the second question is generated. The first question cluster is associated with at least one of the first answer and the second answer. Finally, upon determining that a number of questions in the plurality of questions that are included in the first question cluster exceeds a first predefined threshold, the first question cluster is ingested into a question answering system.

Abstract: A spoken language understanding apparatus according to embodiments of the present disclosure may include: a slot tagging module including: a morpheme analysis unit configured to analyze morphemes with respect to an uttered sentence, a slot tagging unit configured to tag slots corresponding to a semantic entity from a plurality of input tokens generated according to the analyzed morphemes, and a slot name conversion unit configured to convert phrases corresponding to the tagged slots into delexicalized slot names based on neighboring contextual information; and a language generation module configured to generate a combined sequence by combining the delexicalized slot names based on the plurality of input tokens.

Abstract: An audio processing device includes a feature extraction unit and signal generating unit. The feature extraction unit is configured to extract a feature quantity of a first audio signal for each of a plurality of periods. The signal generating unit is configured to for generate a second audio signal by time axis expanding/compressing either a section of the first audio signal in which the feature quantity is steadily maintained for a period time, or a section of the first audio signal in which a fluctuation of the feature quantity is repeated and excluding from the time axis expanding/compressing a section of the first audio signal in which a fluctuation of the feature quantity is not similar to that of other sections of the first audio signal.

Abstract: A speech recognition method to be performed by a computer, the method including: detecting a first keyword uttered by a user from an audio signal representing voice of the user; detecting a term indicating a request of the user from sections that follow the first keyword in the audio signal; and determining a type of speech recognition processing applied to the following sections in accordance with the detected term indicating the request of the user.

Abstract: A keyword extraction method is provided. A candidate keyword from target text is extracted by a server. For each candidate keyword, each effective feature corresponding to the candidate keyword is obtained by the server. Calculation is performed by the server according to each effective feature corresponding to the candidate keyword and a weighting coefficient respectively corresponding to each effective feature, to obtain a probability that the candidate keyword belongs to a target keyword, and the candidate keyword is determined as the target keyword of the target text based on the probability.

Abstract: The present disclosure involves systems, software, and computer implemented methods for personalizing interactions within a conversational interface based on an input context. One example system performs operations including receiving a conversational input via a conversational interface associated with a particular user profile. The input is analyzed via a natural language processing engine to determine an intent and a personality input type. A persona response type associated with the determined personality input type is identified, and responsive content is determined. A particular persona associated with the particular user profile based on a related set of social network activity information associated with the user profile and that corresponds to the identified persona response type is identified.

Abstract: A computer implemented method includes building a Positive Knowledge Base with directive words, designated verbs and designated objects. A Negative Knowledge Base with designated phrases and designated legal terms is built. Tasks and phrases from the Positive Knowledge Base and the Negative Knowledge Base are built. Regulations are received. Phrases from the regulations are weighted against the Positive Knowledge Base and the Negative Knowledge Base to isolate positive Maintenance Compliances. The positive Maintenance Compliances are matched to tasks to derive ranked Maintenance Compliances. The ranked Maintenance Compliances are supplied.

Abstract: An example method includes receiving, by a computational assistant executing at one or more processors, a representation of an utterance spoken at a computing device; selecting, based on the utterance, an agent from a plurality of agents, wherein the plurality of agents includes one or more first party agents and a plurality of third-party agents; responsive to determining that the selected agent comprises a first party agent, selecting a reserved voice from a plurality of voices; and outputting synthesized audio data using the selected voice to satisfy the utterance.

Abstract: An example method includes receiving, by a computational assistant executing at one or more processors, a representation of an utterance spoken at a computing device; identifying, based on the utterance, a task to be performed; determining a capability level of a first party (1P) agent to perform the task; determining capability levels of respective third party (3P) agents of a plurality of 3P agents to perform the task; responsive to determining that the capability level of the 1P agent does not satisfy a threshold capability level, that a capability level of a particular 3P agent of the plurality of 3P agents is a greatest of the determined capability levels, and that the capability level of the particular 3P agent satisfies the threshold capability level, selecting the particular 3P agent to perform the task; and performing one or more actions determined by the selected agent to perform the task.

Abstract: An electronic device includes a controller and a communication device. The controller acquires data indicating first text in a first language. The controller determines whether or not a secret character string is included in the first text. Upon determining that the secret character string is included in the first text, the controller converts the secret character string into a mask character string. The mask character string is for hiding the secret character string. The controller transmits data indicating first text including the mask character string to a translation server through the communication device. The translation server translates the first text into second text in a second language. When the communication device receives data indicating the second text, the controller searches the second text for a translated mask character string. The controller converts the mask character string in the second language into a secret character string in the second language.

Abstract: Techniques for dynamically maintaining speech processing data on a local device for frequently input commands are described. One or more devices receive speech processing data specific to one or more commands associated with system input frequencies satisfying an input frequency threshold. The device(s) then receives input audio corresponding to an utterance and generate input audio data corresponding thereto. The device(s) performs speech recognition processing on input audio data to generate input text data using a portion of the received speech processing data. The device(s) determines a probability score associated with the input text data and determines the probability score satisfies a threshold probability score. The device(s) then performs natural language processing on the input text data to determine the command using a portion of the speech processing data. The device(s) then outputs audio data responsive to the command.

Abstract: An information processing device acts as a neural network based on time-series data. The information processing device includes a memory and a processor. The memory stores an input variable having an ordinal number in the time-series data, and a parameter group for the neural network. The processor calculates an intermediate variable for each ordinal number based on the input variable having the ordinal number by performing transformation based on the parameter group, and calculates an output variable having the ordinal number based on the calculated intermediate variable. Upon calculating an (n+1)-th intermediate variable, the processor performs weighted sum of a calculation result of an n-th intermediate variable and a transformation result in which the n-th intermediate variable and an (n+1)-th input variable are transformed based on the parameter group, to calculate the (n+1)-th intermediate variable.

Abstract: A text-to-speech method includes outputting an instruction according to voice information entered by a user; obtaining text information according to the instruction; converting the text information to audio; and playing the audio. According to the embodiments of the present invention, news or other text content in a browser can be played by voice, which liberates hands and eyes of a user. The user can use the browser in some scenarios where the user cannot easily use the browser, such as driving a car, thereby improving user experience.

Abstract: Embodiments for recommending predictive modeling methods and features by a processor. One or more extracted methods and features of one or more predictive models are received according to selected criteria from both a structured database and from one or more data sources from a remote database. One or more extracted predictive model methods and features may be recommended according to the selected criteria.

Abstract: Methods, systems, and computer-readable storage media for receiving a source domain data set including a set of source document and source label pairs, each source label corresponding to a source domain and indicating a sentiment attributed to a respective source document, receiving a target domain data set including a set of target documents absent target labels, processing documents of the source and target domains using a feature encoder of a DAS platform, to map the documents of the source and target domains to a shared feature space through feature representations, the processing including minimizing a distance between the feature representations of the source domain, and feature representations of the target domain based on a set of loss functions, providing an ensemble prediction from the processing, and providing predicted labels based on the ensemble prediction, the predicted labels being used by the sentiment classifier to classify documents from the target domain.

Abstract: Systems and methods for a speech recognition system for recognizing speech including overlapping speech by multiple speakers. The system including a hardware processor. A computer storage memory to store data along with having computer-executable instructions stored thereon that, when executed by the processor is to implement a stored speech recognition network. An input interface to receive an acoustic signal, the received acoustic signal including a mixture of speech signals by multiple speakers, wherein the multiple speakers include target speakers. An encoder network and a decoder network of the stored speech recognition network are trained to transform the received acoustic signal into a text for each target speaker. Such that the encoder network outputs a set of recognition encodings, and the decoder network uses the set of recognition encodings to output the text for each target speaker. An output interface to transmit the text for each target speaker.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output sequence of audio data that comprises a respective audio sample at each of a plurality of time steps. One of the methods includes, for each of the time steps: providing a current sequence of audio data as input to a convolutional subnetwork, wherein the current sequence comprises the respective audio sample at each time step that precedes the time step in the output sequence, and wherein the convolutional subnetwork is configured to process the current sequence of audio data to generate an alternative representation for the time step; and providing the alternative representation for the time step as input to an output layer, wherein the output layer is configured to: process the alternative representation to generate an output that defines a score distribution over a plurality of possible audio samples for the time step.