Abstract: A method for use with a computing device is provided. The method may include executing one or more programs of an intelligent digital assistant system at a processor and presenting a user interface to a user. At the processor, the method may include receiving natural language user input from the user, parsing the user input at an intent handler to determine an intent template with slots, populating the slots in the intent template with information from user input, and performing resolution on the intent template to partially resolve unresolved information. If a slot with missing slot information exists in the partially resolved intent template, a loop may be executed at the processor to fill the slots. The method may include, at the processor, determining that all required information is available and resolved and generating a rule based upon the intent template with all required information being available and resolved.

Abstract: A system of determining identity based on voiceprint and voice password, and a method thereof are disclosed. In the method, after the voice signal is received, the judgment result of the voiceprint of the voice signal and the judgment result of the content of the voice signal are used to determine whether to pass the verification, and this technical solution of the present invention can confirm that the voice, identified based on the voiceprint thereof, is made by a real person, so as to improve the security of identity determination.

Abstract: Provided are systems, computer-implemented methods, and computer-program products for a multi-lingual device, capable of receiving verbal input in multiple languages, and further capable of providing conversational responses in multiple languages. In various implementations, the multi-lingual device includes an automatic speech recognition engine capable of receiving verbal input in a first natural language and providing a textual representation of the input and a confidence value for the recognition. The multi-lingual device can also include a machine translation engine, capable of translating textual input from the first natural language into a second natural language. The machine translation engine can output a confidence value for the translation. The multi-lingual device can further include natural language processing, capable of translating from the second natural language to a computer-based language.

Abstract: An email-like user interface displays a list of user logs determined based on user-specified list criteria to user logs received in a natural language (NL) training environment. The list comprise a subset of the received user logs in order to minimize the number of actions required to configure and train the NL configuration system in a semi-supervised manner, thereby improving the quality and accuracy of NL configuration system. To determine a list of user logs relevant for training the user logs can be filtered, sorted, grouped and searched within the email-like user interface. A training interface to a network of instances that comprises a plurality of NL configuration systems leverages a crowd-sourcing community of developers in order to efficiently create a customizable NL configuration system.

Abstract: The present disclosure is directed to systems and methods for the creation of a localized audio message for use in a personal audio device. The system includes: a database of information relating to a pre-determined subject obtained from online media content; one or more processors; and a personal audio device configured to receive a localized audio message. The processors extract a dataset comprising information relating to a pre-determined subject from online media content; generate one or more summaries of the information relating to the pre-determined subject; generate a localized audio message based on the one or more summaries; and send the localized audio message to a personal audio device of a user.

Abstract: A word score calculation device according to an embodiment includes an input unit and a calculating unit. The input unit receives input of a string of words including known words, which are known already, and new words, which are not yet known. The calculating unit inputs the lexicon feature of a word, which is included in the string of words, in a neural network in which one or more unit groups for inputting the lexicon features of words are included in an input layer, and calculates a word score of the target word included in the string of words.

Abstract: Provided are an image display apparatus and a method of controlling the same. The image display apparatus enabling voice recognition includes: a first voice inputter which receives a user-side audio signal; an audio outputter which outputs an audio signal processed by the image display apparatus; a first voice recognizer which recognizes the user-side audio signal received through the first voice inputter; and a controller which decreases a volume of the audio signal output through the audio outputter to a predetermined level if a voice recognition start command is received.

Abstract: Embodiments include computer-implemented methods and systems for detecting undesirable and potentially harmful online behavior. The embodiments described and claimed could also be applied to detecting any other type of online behavior to be detected, but the descriptions focuses on detecting online violence. More particularly, the embodiments disclosed relate to detecting online violence using symbolic methods of natural language processing (NLP) that utilize and govern the usage of: 1) syntactic parser for analyzing grammatical context of the input text data, 2) unsupervised learning methods for improving selected aspects of the system and adjusting the system to new data sources and guidelines, and 3) statistical classifiers for resolving specific well-defined sub-tasks, in which statistical approaches surpass the symbolic methods.

Abstract: A mechanism, method, and computer readable medium to enhance speech enabled devices. The method comprising receiving, by an ultrasonic attack detector of a speech enabled device, an audio stream from one or more microphones and a segmentation signal from a keyword detector indicating a location of a detected keyword within the audio stream, preprocessing, by the ultrasonic attack detector, a segmented portion of the audio stream including the detected keyword to obtain a spectrogram, and executing, by the ultrasonic attack detector, a neural network classifier using the spectrogram as input, the neural network classifier to discern real human speech from intermodulation distortion products resulting from ultrasonic attacks on the speech enabled device.

Abstract: An input audio signal comprises a plurality of input audio channels. A KLT-based pre-processor transforms the plurality of input audio channels into a plurality of eigenchannels and provides metadata associated with the plurality of eigenchannels. Each eigenchannel is associated with an eigenvalue and an eigenvector. The metadata allows reconstructing the plurality of input audio channels on the basis of the plurality of eigenchannels. A selector selects a subset of the plurality of eigenvectors corresponding to a plurality of selected eigenchannels on the basis of a geometric mean of the eigenvalues. An eigenchannel encoder encodes the plurality of selected eigenchannels. A metadata encoder encodes the metadata.

Abstract: Methods, systems, and apparatus, including computer programs stored on computer storage devices, for multi-tiered command processing are disclosed. One of the methods includes operating an audio processing system to detect voice commands in a set of first commands controlling behavior of the robot and in a set of second commands controlling behavior of the robot, wherein the system detects commands in the set of first commands using a local recognition model that does not require communication over a network, and detects commands in the set of second commands with a server over a network; executing one or more applications to respond to commands in a set of application commands that include commands different from the first commands and second commands; and detecting and carrying out commands, according to a predetermined hierarchy that prioritizes detection of the first commands first, the second commands second, and the application commands third.

Abstract: Techniques for script modification are provided including receiving a script and parsing the script to identify at least one attribute of the script. The identified at least one attribute is presented to a user in a graphical user interface via a display device and an adjustment of at least one element in the graphical user interface that corresponds to the at least one attribute is received from the user via an input device. Modification data corresponding to the at least one attribute are received from a data repository and at least one attribute of the script is modified based on the received adjustment and the obtained modification data corresponding to the at least one attribute. A modified script is generated based on the modified at least one attribute.

Abstract: There are provided mechanisms for determining an uncertainty measure of a mixture-model based parametric classifier. A method is performed by a classification device. The method includes obtaining a short-term frequency representation of a multimedia signal. The short-term frequency representation defines an input sequence. The method includes classifying the input sequence to belong to one class of at least two available classes using the parametric classifier. The parametric classifier has been trained with a training sequence. The method includes determining an uncertainty measure of the classified input sequence based on a relation between posterior probabilities of the input sequence and posterior probabilities of the training sequence.

Abstract: An apparatus for generating an enhanced signal from an input signal, wherein the enhanced signal has spectral values for an enhancement spectral region, the spectral values for the enhancement spectral regions not being contained in the input signal, includes a mapper for mapping a source spectral region of the input signal to a target region in the enhancement spectral region, the source spectral region including a noise-filling region; and a noise filler configured for generating first noise values for the noise-filling region in the source spectral region of the input signal and for generating second noise values for a noise region in the target region, wherein the second noise values are decorrelated from the first noise values or for generating second noise values for a noise region in the target region, wherein the second noise values are decorrelated from first noise values in the source region.

Abstract: A wearable device providing an augmented reality experience for the benefit of hearing impaired persons is disclosed. The augmented reality experience displays a virtual text caption box that includes text that has been translated from speech detected from surrounding speakers.

Abstract: The present technology reduces a process load in a reception side when a plurality of types of audi data is transmitted. A metafile having meta information used to acquire, in a reception device, a predetermined number of audio streams including a plurality of groups of encoded data is transmitted. To the metafile, attribute information indicating each attribute of the encoded data of the plurality of groups is inserted. For example, to the metafile, stream correspondence relation information indicating in which audio stream the encoded data of the plurality of groups is included respectively is further inserted.

Abstract: Disclosed are a packet loss concealment method and apparatus a using a generative adversarial network. A method for packet loss concealment in voice communication may include training a classification model based on a generative adversarial network (GAN) with respect to a voice signal including a plurality of frames, training a generative model having a contention relation with the classification model based on the GAN, estimating lost packet information based on the trained generative model with respect to the voice signal encoded by a codec, and restoring a lost packet based on the estimated packet information.

Abstract: In order to detect a replay attack in a speaker recognition system, at least one feature is identified in a detected magnetic field. It is then determined whether the at least one identified feature of the detected magnetic field is indicative of playback of speech through a loudspeaker. If so, it is determined that a replay attack may have taken place.

Abstract: There is provided an information processing device, an information processing method, and a program which are capable of performing voice recognition adaptively to the degree of excitement in the sound collection state. The information processing device includes: an acquiring unit configured to acquire information indicating a degree of excitement in a collection state of a voice; and a voice recognizing unit configured to perform first voice recognition based on a phoneme of the voice on the basis of the information indicating the degree of excitement.

Abstract: Embodiments of the invention are directed to systems and methods for biometrics transaction processing. A location of a device associated with a user may be determined. A reference to a biometric data model associated with the user stored within a database may be retrieved, based at least in part on the location. Biometric data may be received from the user. Using the reference, the biometric data may be compared to the biometric data model stored within the database. A determination may be made whether the user is authenticated for the transaction based on the comparing step.