Abstract: Provided is an information processing device, an information processing method, and a program, the information processing device including a control unit that dynamically controls output of notification information related to a function corresponding to a gesture regarding function execution of the device based on a recognition status of an operation body that is executing the gesture in a predetermined operation region.

Abstract: Described herein are system and method embodiments to improve word representation learning. Embodiments of a probabilistic prior may seamlessly integrate statistical disentanglement with word embedding. Different from previous deterministic methods, word embedding may be taken as a probabilistic generative model, and it enables imposing a prior that may identify independent factors generating word representation vectors. The probabilistic prior not only enhances the representation of word embedding, but also improves the model's robustness and stability. Furthermore, embodiments of the disclosed method may be flexibly plugged in various word embedding models. Extensive experimental results show that embodiments of the presented method may improve word representation on different tasks.

Abstract: Media unit retrieval methods, systems and computer program products are provided that allow a user to search for an item by iteratively presenting media units such as images representing items to the user and receiving user input consisting of selections of the presented media units (including possibly the empty selection). Features, or attributes, a user is interested in, for example semantic features, are inferred from the interaction and media units are retrieved for presentation based on similarity with user-selected media units, through sampling of a probability distribution describing the intent or interests, or combinations of approaches. Accordingly, the user-experience is akin to a conversation about what the user is looking for. Retrieval may be based on both selected and unselected media units and the selection may comprise making a selection with a single action. Further, a database of media units can capture similarity relationships for efficient media unit retrieval.

Abstract: A boundary of a highlight of audiovisual content depicting an event is identified. The audiovisual content may be a broadcast, such as a television broadcast of a sporting event. The highlight may be a segment of the audiovisual content deemed to be of particular interest. Audio data for the audiovisual content is stored, and the audio data is automatically analyzed to detect one or more audio events indicative of one or more occurrences to be included in the highlight. Each audio event may be a brief, high-energy audio burst such as the sound made by a tennis serve. A time index within the audiovisual content, before or after the audio event, may be designated as the boundary, which may be the beginning or end of the highlight.

Abstract: Various embodiments of a system and associated method for anonymization of text without losing semantic utility of text by extracting a latent embedding representation of content with respect to a given task and by learning an optimal strategy for text embedding manipulation to satisfy both privacy and utility requirements are disclosed herein. In particular, the system balances private attribute obfuscation with retained semantic utility.

Abstract: Methods and devices for enabling and disabling applications using voice are described herein. In some embodiments, an individual speak an utterance to their electronic device, which may send audio data representing the utterance to a backend system. The backend system may generate text data representing the utterance, and may determine that an intent of the utterance was for an application to be enabled or disabled for their user account on the backend system. If, for instance, the intent was to enable the application, the backend system may receive one or more rules for performing functionalities of the application, as well as one or more sample templates of sample utterances and sample responses that future utterances may use when requesting the application. Furthermore, one or more invocation phrases that may be used within the future utterances to invoke the application may be received, along with slot values for the sample templates.

Abstract: A configurable input element of a controlling device is configured by using a data representative of an over-the-top (OTT) media app determined to be installed on an OTT device and a data representative of the OTT device to identify at least one command that is required to be transmitted to cause the OTT device to launch the OTT media app. The at least one command is provisioned to the controlling device and assigned to the configurable input element. When the input element is subsequently activated, the controlling device will transmit the at least one command to cause the OTT device to launch the OTT media app.

Abstract: Methods, systems and computer program products for automated learning are provided herein. A computer-implemented method includes authenticating a plurality of users for an automated learning session, wherein the plurality of users correspond to at least one device, and providing the automated learning session for the plurality of users. Providing the automated learning session comprises analyzing a plurality of learning models corresponding to one or more of the plurality of users, determining, based on the analysis, one or more activities to be performed by the plurality of users during the automated learning session, and executing the one or more activities on at least one device.

Abstract: A parameter included in a fundamental frequency pattern of a voice can be estimated from the fundamental frequency pattern with high accuracy and the fundamental frequency pattern of the voice can be reconstructed from the parameter included in the fundamental frequency pattern. A learning unit 30 learns a deep generation model including an encoder which regards a parameter included in a fundamental frequency pattern in a voice signal as a latent variable of the deep generation model and estimates the latent variable from the fundamental frequency pattern in the voice signal on the basis of parallel data of the fundamental frequency pattern in the voice signal and the parameter included in the fundamental frequency pattern in the voice signal, and a decoder which reconstructs the fundamental frequency pattern in the voice signal from the latent variable.

Abstract: Embodiments described herein provide adversarial attacks targeting the cross-lingual generalization ability of massive multilingual representations, demonstrating their effectiveness on multilingual models for natural language inference and question answering. An efficient adversarial training scheme can thus be implemented with the adversarial attacks, which takes the same number of steps as standard supervised training and show that it encourages language-invariance in representations, thereby improving both clean and robust accuracy.

Abstract: The present invention discloses a method and apparatus for awakening skills by speech, which are applied to an electronic device. The method for awakening skills by speech includes: recognizing awakening text information corresponding to a speech request message to be processed; invoking a service skill semantic model to determine a target service field corresponding to the awakening text information and a corresponding first confidence, and invoking a knowledge skill semantic model to determine a knowledge reply answer corresponding to the awakening text information and a corresponding second confidence; and selecting to awaken one of a knowledge skill and a target service skill corresponding to the target service field based on the first confidence and the second confidence. Accordingly, the probability of erroneously awakening a skill based on the speech message can be reduced.

Abstract: The following relates generally to voice assisted healthcare. In some embodiments, a digital assistant receives audio data, and determines an intent from the audio data. The digital assistant may then match the determined intent to a flow of a set of flows, where the set of flows may include at least one of: (i) submitting a prescription, (ii) refilling a prescription, (iii) changing a pickup location, (iv) requesting a status update for a prescription, or (v) initiating a pharmacy chat session. The matched flow of the set of flows may then be executed.

Abstract: A method and an apparatus for enabling adaptive audio signal alteration are described. When an input audio signal is received, a determination of whether the user of an audio device hears the input audio signal is performed based upon brain activity of the user. A determination of whether the user is distracted by the audio signal is performed based upon sensor measurements indicating a physical state of the user. In response to determining that the user hears the input audio signal and that the input audio signal causes the user to be distracted, a determination of configuration parameter(s) is performed. An alteration of audio signal(s) is caused based upon the configuration parameter(s) to obtain modified version(s) of the audio signal(s) that are intended to address the distraction caused by the input audio signal, and output audio signals are output, where the output audio signals include the modified versions.

Abstract: Techniques for implementing a “sticky” user ID are described. A system receives first input audio data and determines first speech processing results therefrom. The system also determines a first user ID of a user that spoke an utterance represented in the first input audio data and associates the first user ID with a device, which originated the first input audio data, for a predetermined length of time. The system determines first output data responsive to the first speech processing data and causes the device to present first output content corresponding thereto. The system then receives second input audio data and determines second speech processing results therefrom. The system also determines a time of receipt of the second input audio data is within the predetermined length of time. Based at least in part thereon, the system determined second output data responsive to the second speech processing data using the first user ID.

Abstract: A new chest X-ray database, referred to as “ChestX-ray8”, is disclosed herein, which comprises over 100,000 frontal view X-ray images of over 32,000 unique patients with the text-mined eight disease image labels (where each image can have multi-labels), from the associated radiological reports using natural language processing. We demonstrate that these commonly occurring thoracic diseases can be detected and spatially-located via a unified weakly supervised multi-label image classification and disease localization framework, which is validated using our disclosed dataset.

Abstract: Methods and apparatus for audio identification during a performance are disclosed herein. An example apparatus includes at least one memory and at least one processor to transform a segment of audio into a log-frequency spectrogram based on a constant Q transform using a logarithmic frequency resolution, transform the log-frequency spectrogram into a binary image, each pixel of the binary image corresponding to a time frame and frequency channel pair, each frequency channel representing a corresponding quarter tone frequency channel in a range from C3-C8, generate a matrix product of the binary image and a plurality of reference fingerprints, normalize the matrix product to form a similarity matrix, select an alignment of a line in the similarity matrix that intersects one or more bins in the similarity matrix with the largest calculated Hamming similarities, and select a reference fingerprint based on the alignment.

Abstract: A configurable input element of a controlling device is configured by using a data representative of an over-the-top (OTT) media app determined to be installed on an OTT device and a data representative of the OTT device to identify at least one command that is required to be transmitted to cause the OTT device to launch the OTT media app. The at least one command is provisioned to the controlling device and assigned to the configurable input element. When the input element is subsequently activated, the controlling device will transmit the at least one command to cause the OTT device to launch the OTT media app.

Abstract: An information processing system includes: an image display apparatus provided in a space and configured to display an image; a sensor apparatus carried by a user who is present in the space and configured to output a signal for detecting position information of the user in the space; and an information processing apparatus. The information processing apparatus includes circuitry configured to store a plurality of pieces of position information of a plurality of users including the user, who are in present in the space, in association with the plurality of users, the plurality of users being detected based on signals output from a plurality of sensor apparatuses including the sensor apparatus, and control environment effect production that supports communication between the plurality of users by the image displayed by the image display apparatus, based on each of the plurality of pieces of position information of the plurality of users.

Abstract: A device with a microphone acquires audio data of a user's speech. A neural network accepts audio data as input and provides sentiment data as output. The neural network is trained using training data based on input from raters who provide votes as to which sentiment descriptors they think are associated with a sample of speech. A vote by a rater assessing the sample for a particular semantic descriptor is distributed to a plurality of semantically similar semantic descriptors. Semantic descriptor similarity data indicates relative similarity between possible semantic descriptors in the semantic space. The distributed partial votes may be aggregated to produce training data comprising samples of speech and weights of corresponding semantic descriptors. The training data is then used to train the neural network. For example, the neural network may be trained with the training data using per-instance cosine similarity loss or correlational loss.

Abstract: A method and system for detecting errors when practicing fluency shaping exercises. The method includes setting each threshold of a set of thresholds to a respective predetermined initial value; analyzing a voice production to compute a set of first energy levels composing the voice production, wherein the voice production is of a user practicing a fluency shaping exercise; detecting at least one speech-related error based on the computed set of first energy levels, a set of second energy levels, and the set of thresholds, wherein the detection of the at least one speech-related error is with respect to the fluency shaping exercise being practiced by the user, wherein the set of second energy levels is determined based on a calibration process; and generating feedback indicating the detected at least one speech-related error.

Abstract: Pressure sensing guidewire assemblies are described herein where the guidewire assembly may be comprised of an elongate guidewire body and multiple pressure sensors secured near or at a distal end of the guidewire body. The signals obtained from the guidewire connectors and aortic sensor modules may be synchronized to minimize signal acquisition delays. The signals may be further processed to equalize the pressure waveforms by shifting the connector waveform to align correctly with the aortic module waveform and improve output signals.

Abstract: A system configured to enable a Wi-Fi processor to enter a low power mode (LPM) for short periods of time without compromising functionality is provided. A device reduces power consumption by enabling the Wi-Fi processor to enter LPM with scheduled wakeup events to enable specific functionality. In some examples, the Wi-Fi processor toggles between LPM and an active mode based on a first duty cycle to enable new device provisioning. The first duty cycle corresponds to a time required to scan a plurality of wireless channels, waking the Wi-Fi processor at a first frequency to monitor for incoming probe requests. In other examples, the Wi-Fi processor uses a second duty cycle chosen to maintain time synchronicity between a time master device and time follower devices. The device sets the second duty cycle to wake the Wi-Fi processor at a second frequency to exchange data packets with synchronized devices.

Abstract: Existence of instrumentation for automatic video recording creates an excess capacity of video recording for those who own automatic video recorders. Others may want to utilize this excess capacity to record their activities thus there is a need for a system that helps match those who would like to utilize the excess capacity with those who have such capacity. Such excess capacity is matched with demand to use such excess capacity by creating a network of automatic video recording units and tags that are associated with people who want to be recorded.

Abstract: The present invention allows a robot to carry out communication with excellent affectiveness. A speech and behavior control device (1) includes an utterance content selecting section (16) which selects utterance content of a robot (100) from among a plurality of utterances, a movement control section (17) which controls a movable part (13) to move based on a kind of feeling corresponding to the utterance content, and an audio control section (18) which controls the robot (100) to output the utterance content as audio after movement of the movable part (13) has been started.

Abstract: The described method and system provide for HMI steering for a telematics-equipped vehicle based on likelihood to exceed eye glance guidelines. By determining whether a task is likely to cause the user to exceed eye glance guidelines, alternative HMI processes may be presented to a user to reduce ASGT and EORT and increase compliance with eye glance guidelines. By allowing a user to navigate through long lists of items through vocal input, T9 text input, or heuristic processing rather than through conventional presentation of the full list, a user is much more likely to comply with the eye glance guidelines. This invention is particularly useful in contexts where users may be searching for one item out of a plurality of potential items, for example, within the context of hands-free calling contacts, playing back audio files, or finding points of interest during GPS navigation.

Abstract: A method is provided for managing phrase completion suggestions in response to text input. The method includes receiving text entered into the computing system, and identifying a first plurality of phrases that each begins with the received text and that each includes a respective phrase segment immediately following the received text. The method further includes displaying a first list of the respective phrase segments of the identified first plurality of phrases without displaying the received text, and receiving input defining a selection of one of the respective phrase segments of the displayed first list.

Abstract: A mobile device moves by calculating a distance between a sound source and the mobile device using a sound source direction estimation technique. The mobile device moves by a reference distance in a direction perpendicular to a direction in which the mobile device faces the sound source when call sound of the sound source is generated, outputs voice to instruct to the sound source to generate recall sound, checks a directional angle of the mobile device when recall sound is generated by the sound source, calculates the distance between the sound source and the mobile device according to the reference distance and the directional angle of the mobile device, and moves to the vicinity of the sound source.

Abstract: A chatbot system and method with contextual input/output messages. A chatbot includes a processor, an interactive dialog interface and a knowledge database. The system uses a script file to display input and output messages in a tree format. An initial input or output message is stored. An identifier is assigned to the initial input or output message that is then used as context for the subsequent input/output messages by associating and storing the identifier with the subsequent input/output messages. The relationship between the first input or output message and subsequent input/output messages define a parent-child relationship that is displayable via the script file.

Abstract: A graphical sketch can be received, the sketch including one or more representations of text. A query can be automatically generated from the sketch. The generation of the query can include automatically recognizing the text and automatically representing the text in the query. The query can be run to identify a picture in response to the query, with the text describing one or more non-textual features of the picture. The picture can be returned, such as in response to the receipt of the graphical sketch.

Abstract: In an embodiment, a system maintains a database of a plurality of persons. The database includes an audio clip of a pronunciation of a name of a first person in the database. The system determines from a calendar database that a second person has an event in common with the first person, and transmits to a device associated with the second person an indication that the database includes the pronunciation of the name of the first person.

Abstract: In-vehicle functions are implemented using a plurality of microphones disposed in a vehicle. Each of the microphones is disposed in a portion of the vehicle defined by a zone. The in-vehicle functions are also implemented via a central controller of the vehicle. The central controller includes a computer processor executing logic. The logic receive a voice communication from an individual via one of the microphones, identifies the zone in the vehicle occupied by the individual, identifies the individual by comparing a voice stamp from the voice communication to a database of voice stamps, and implements at least one vehicle electronic component in the zone based on user preferences associated with the voice stamp.

Abstract: A control method for a video-audio playing system receiving a video-audio streaming signal is provided. The video-audio streaming signal includes at least a channel-program information. The control method comprises receiving a speech signal and analyzing the speech signal to obtain an acoustic feature of the speech signal. According to the acoustic feature, a speech recognition is performed to determine one of the channel-program information corresponds to the acoustic feature. According to the determined channel-program information, the video-audio playing system executes an operation corresponding to the channel-program information.

Abstract: A medical error alert device may comprise a controller; a first memory, a recording and playback module and a user interface. The user interface may be configured to enable a patient or a patient representative to record an announcement identifying at least a medical procedure to be carried out. The user interface may be further configured to enable later playback of the announcement before the medical procedure is carried out. A communication device may be provided, coupled to a network to enable reception of signals from the network comprising at least predetermined patient identification number and/or a unique medical alert device identifier. A predetermined alert may be generated responsive to the communication device receiving a signal associated with the predetermined alert and the patient identification number and/or the unique device identifier.

Abstract: The embodiments disclosed herein relate to a system and method for processing a prescription through voice-activated commands. The system and method efficiently and effectively process the prescription so that a pharmacy may handle the increasing prescription processing demands.

Abstract: Dynamic adjustment of text input system components is provided. An indication of user activity with respect to a text input system of an electronic device is received. One or more activity indicators are determined based on at least the user activity. One or more components of the text input system are identified, each component providing a typing assistance functionality to a user and being associated with a set of parameters. For each of the one or more components, a determination is made whether the component should be adjusted based on the one or more activity indicators, and the component is dynamically adjusted when it is determined that the component should be adjusted based on the one or more activity indicators. Dynamically adjusting the component includes at least one of activating the component, deactivating the component or adjusting the set of parameters associated with the component.

Abstract: A medical error alert device may comprise a controller; a first memory, a recording and playback module and a user interface. The user interface may be configured to enable a patient or a patient representative to record an announcement identifying at least a medical procedure to be carried out. The user interface may be further configured to enable later playback of the announcement before the medical procedure is carried out. A communication device may be provided, coupled to a network to enable reception of signals from the network comprising at least predetermined patient identification number and/or a unique medical alert device identifier. A predetermined alert may be generated responsive to the communication device receiving a signal associated with the predetermined alert and the patient identification number and/or the unique device identifier.

Abstract: In one embodiment, a method comprises classifying a representation of audio data of a dialog turn in a dialog system to a classification. The method may further comprise taking a security action on the classified representation of the audio data of the dialog turn as a function of the classification. The security action can be suppressing the representation of the audio data, encrypting the representation of the audio data, releasing the representation of the audio data, partially suppressing the representation of the audio data, partially encrypting the representation of the audio data, partially releasing the representation of the audio data, or a command.

Abstract: A dialog system is accessed by a remote user and is typically configured to receive a natural language query from the user and return a natural language answer to the user. Dialog systems can be copied without authorization or can become an out-of-date version. A dialog system with a signature, referred to herein as a “signed” dialog system, can indicate the signature without affecting usage by users who are unaware that the dialog system contains the signature. The signed dialog system can respond to input such that only the designer of the dialog system knows the signature is embedded in the dialog system. The response is a way to check the source or other characteristics of the dialog system. A designer of signed dialog systems can prove whether an unauthorized copy of the signed dialog system is used by a third party by using publically-available user interfaces.

Abstract: A voice activation system is provided. The voice activation system includes a first stage configured to output a first activation signal if at least one energy characteristic of a received audio signal satisfies at least one threshold and a second stage configured to transition from a first state to a second state in response to the first activation signal and, when in the second state, to output a second activation signal if at least a portion of a profile of the audio signal substantially matches at least one predetermined profile.

Abstract: A remote controller includes a housing, a direction sensor, a microphone, a controller, and a wireless transmitter. A control method of the remote controller includes detecting an angle between an axis of a remote controller and a vertical axis, enabling a microphone of the remote controller when the angle is within a predetermined range in order to generate a voice signal according to a voice command, and generating a first control signal according the voice signal and transmit the first control signal wirelessly.

Abstract: The instant application includes computationally-implemented systems and methods that include acquiring indication of a speech-facilitated transaction between a particular party and a target device, receiving adaptation data correlated to the particular party, the receiving facilitated by a particular device associated with the particular party, processing audio data from the particular party at least partly using the received adaptation data correlated to the particular party, and updating the adaptation data based at least in part on a result of the processed audio data, such that the updated adaptation data is configured to be transmitted to the particular device. In addition to the foregoing, other aspects are described in the claims, drawings, and text.

Abstract: The instant application includes computationally-implemented systems and methods that include acquiring indication of a speech-facilitated transaction between a particular party and a target device, receiving adaptation data correlated to the particular party, the receiving facilitated by a particular device associated with the particular party, processing audio data from the particular party at least partly using the received adaptation data correlated to the particular party, and updating the adaptation data based at least in part on a result of the processed audio data, such that the updated adaptation data is configured to be transmitted to the particular device. In addition to the foregoing, other aspects are described in the claims, drawings, and text.

Abstract: Disclosed is an on demand, web-based, outbound contact center utilizing Voice over IP (VoIP) and speaker-independent voice recognition which automatically captures contact responses to question events in a pre-recorded, interactive voice call, the call launched by a user via a broadcast comprising a call sequence created by the user via a call center user interface comprising event add and logic add wizards, the call sequence comprising event prompts based on a user-generated script comprising message events and question events, the event prompts in the group consisting of voice recordings and text-to-speech inputs.

Abstract: A computer-implemented system includes one or multiple application devices and a voice-controlled storage device. Multiple voice commands may be issued to multiple application devices simultaneously or separately, or to the same application device separately. The voice-controlled storage device is configured to perform content identification and voiceprint recognition on the voice commands. Therefore, each requestor may be allowed to operate the voice-controlled storage device in a corresponding operation mode according to respective authorization level.

Abstract: A method of configuring an acoustics system of a convertible vehicle to receive speech from an occupant of the vehicle who is using hand-free technology. The position of the top of the convertible is first determined and based upon whether the top is up or down, an audio reception configuration is selected. The audio reception configuration includes a set of tuning parameters and a microphone arrangement. The acoustics system is then configured based upon the determination of whether the top is up or down.

Abstract: Techniques disclosed herein include systems and methods that enable a voice trigger that wakes-up an electronic device or causes the device to make additional voice commands active, without manual initiation of voice command functionality. In addition, such a voice trigger is dynamically programmable or customizable. A speaker can program or designate a particular phrase as the voice trigger. In general, techniques herein execute a voice-activated wake-up system that operates on a digital signal processor (DSP) or other low-power, secondary processing unit of an electronic device instead of running on a central processing unit (CPU). A speech recognition manager runs two speech recognition systems on an electronic device. The CPU dynamically creates a compact speech system for the DSP. Such a compact system can be continuously run during a standby mode, without quickly exhausting a battery supply.

Abstract: A system and methodology of delivery fluids and monitoring their status which is voice actuated. This system has application where a hands-free environment is preferred. Voice commands are given by the user via a Bluetooth® headset and received typically by the user's Smartphone. Voice recognition circuitry is programmed to recognize the simple commands and through complementing electronics, and electro-mechanical and mechanical elements, delivery at corresponding flow rates is accomplished. A further feature allows for respective voice commands to initiate a monitoring function where the status of any particular characteristic of the fluid can be relayed back to the user via the headset.

Abstract: A procurement system may include a first interface configured to receive a query from a user, a command module configured to parameterize the query, an intelligent search and match engine configured to compare the parameterized query with stored queries in a historical knowledge base and, in the event the parameterized query does not match a stored query within the historical knowledge base, search for a match in a plurality of knowledge models, and a response solution engine configured to receive a system response ID from the intelligent search and match engine, the response solution engine being configured to initiate a system action by interacting with sub-system and related databases to generate a system response.

Abstract: The text-to-speech audio HIP technique described herein in some embodiments uses different correlated or uncorrelated words or sentences generated via a text-to-speech engine as audio HIP challenges. The technique can apply different effects in the text-to-speech synthesizer speaking a sentence to be used as a HIP challenge string. The different effects can include, for example, spectral frequency warping; vowel duration warping; background addition; echo addition; and varying the time duration between words, among others. In some embodiments the technique varies the set of parameters to prevent using Automated Speech Recognition tools from using previously used audio HIP challenges to learn a model which can then be used to recognize future audio HIP challenges generated by the technique. Additionally, in some embodiments the technique introduces the requirement of semantic understanding in HIP challenges.

Abstract: A language interpretation system receives a request for an interpretation of a voice communication between a first language and a second language. Further, the language interpretation system provides the request to a machine language interpreter. In addition, the machine language interpreter provides live language interpretation of the voice communication. The live language interpretation of the voice communication is halted by the machine language interpreter in real time during the live language interpretation based upon a criteria being met. Further, the voice communication is transitioned to a human language interpreter to resume the live language interpretation of the voice communication after the machine language interpreter is halted.