Abstract: A task-oriented dialog system determines an endpoint in a user utterance by receiving incremental portions of a user utterance that is provided in real time during a task-oriented communication session between a user and a virtual agent (VA). The task-oriented dialog system recognizes words in the incremental portions using an automated speech recognition (ASR) model and generates semantic information for the incremental portions of the utterance by applying a natural language processing (NLP) model to the recognized words. An acoustic-prosodic signature of the incremental portions of the utterance is generated using an acoustic-prosodic model. The task-oriented dialog system can generate a feature vector that represents the incrementally recognized words, the semantic information, the acoustic-prosodic signature, and corresponding confidence scores of the model outputs. A model is applied to the feature vector to identify a likely endpoint in the user utterance.

Abstract: System and method are provided to detect objects in a scene frame of two-dimensional (2D) video using image processing and determine object image coordinates of the detected objects in the scene frame. The system and method deploy a virtual camera in a three-dimensional (3D) environment to create a virtual image frame in the environment and generate a floor in the environment in a plane below the virtual camera. The system and method adjust the virtual camera to change a height and angle relative to the virtual image frame. The system and method generate at an extended reality (XR) coordinate location relative to the floor for placing the detected object in the environment. The XR coordinate location is a point of intersection of a ray cast of the virtual camera through the virtual frame on the floor that translates to the image coordinate in the virtual image frame.

Abstract: A natural language understanding (NLU) system generates in-place annotations for natural language utterances or other types of time-based media based on stand-off annotations. The in-place annotations are associated with particular sub-sequences of an annotation, which provides richer information than stand-off annotations, which are associated only with an utterance as a whole. To generate the in-place annotations for an utterance, the NLU system applies an encoder network and a decoder network to obtain attention weights for the various tokens within the utterance. The NLU system disqualifies tokens of the utterance based on their corresponding attention weights, and selects highest-scoring contiguous sequences of tokens between the disqualified tokens. In-place annotations are associated with the selected sequences.

Abstract: Structure of conversations between users and agents and/or systems is discovered and interactively displayed to analysts, thereby better supporting development of automated conversation handling systems for different domains. A corpus of prior dialogs of users with agents (without preexisting semantic labels indicating purposes for different parts of the dialogs) is taken as input, and embeddings are generated for textual units (e.g., rounds) of the dialogs. The embeddings are used to cluster the textual units, and the clusters and their relationships are visualized within a user interface that analysts may use to explore and fine-tune the structure of the conversations.

Abstract: Content management for virtual tours including exposing, within a virtual tour, a hotspot, the hotspot having an associated a hyperlink to a third-party content management provider and access code for accessing content from third-party content management provider; receiving an invocation of the hotspot; invoking the hyperlink associated with the hotspot; invoking the access code for accessing content from third-party content management provider; receiving, from the third-party content management provider, content; and displaying the received content within the virtual tour.

Abstract: A dampener for a banjo that attenuates unwanted overtones/reverberation while preserving the all-important fundamental sound with minimal reduction in bass, treble and volume. The dampener has a cylindrical outer shell having a diameter of about two inches and a length corresponding to the inner arc of the banjo. The cylindrical outer shell is formed from a cotton poplin fabric, and the filling comprises polyester fibers.

Abstract: A speech recognition system includes, or has access to, conventional speech recognizer data, including a conventional acoustic model and pronunciation dictionary. The speech recognition system generates restructured speech recognizer data from the conventional speech recognizer data. When used at runtime by a speech recognizer module, the restructured speech recognizer data produces more accurate and efficient results than those produced using the conventional speech recognizer data. The restructuring involves segmenting entries of the conventional pronunciation dictionary and acoustic model according to their constituent phonemes and grouping those entries with the same initial N phonemes, for some integer N (e.g., N=3), and deriving a restructured dictionary with a corresponding semi-word acoustic model for the various grouped entries.

Abstract: A virtual assistant system for communicating with customers uses human intelligence to correct any errors in the system AI, while collecting data for machine learning and future improvements for more automation. The system may use a modular design, with separate components for carrying out different system functions and sub-functions, and with frameworks for selecting the component best able to respond to a given customer conversation.

Abstract: A touchless gaming system includes one or more gaming stations, a registration device at each gaming station, and a game processor, as well as a personal electronic device used by each participant in a game offered by the touchless gaming system. The touchless gaming system enables an individual to participate in a live game (e.g., within a casino, etc.) while socially distancing the participants and eliminating the need for physical contact with objects that could infections agents. Touchless gaming methods are also disclosed.

Abstract: A virtual assistant system for communicating with customers uses human intelligence to correct any errors in the system AI, while collecting data for machine learning and future improvements for more automation. The system may use a modular design, with separate components for carrying out different system functions and sub-functions, and with frameworks for selecting the component best able to respond to a given customer conversation. The system may have agent assistance functionality that uses natural language processing to identity concepts in a user conversation and to illustrate that concepts within a graphical user interface of a human agent so that the human agent can more accurately and more rapidly assist the user in accomplishing the user's conversational objectives.

Abstract: Within an environment in which users converse at least partly with human agents to accomplish a desired task, a server assists the agents by identifying workflows that are most applicable to the current conversation. Workflow selection functionality identifies one or more candidate workflows based on techniques such as user intent inference, conversation state tracking, or search, according to various embodiments. The identified candidate workflows are either automatically selected on behalf of the agent, or are presented to the agent for manual selection.

Abstract: Systems and methods are disclosed herein for discerning aspects of user speech to determine user intent and/or other acoustic features of a sound input without the use of an ASR engine. To this end, a processor may receive a sound signal comprising raw acoustic data from a client device, and divides the data into acoustic units. The processor feeds the acoustic units through a first machine learning model to obtain a first output and determines a first mapping, using the first output, of each respective acoustic unit to a plurality of candidate representations of the respective acoustic unit. The processor feeds each candidate representation of the plurality through a second machine learning model to obtain a second output, determines a second mapping, using the second output, of each candidate representation to a known condition, and determines a label for the sound signal based on the second mapping.

Abstract: An immersive training system, and method performed with such system, which comprises one or more sensors configured to measure at least one response of a user to a three-dimensional (3D) asset of a 3D rendering that is being rendered on a virtual reality (VR) display, the 3D rendering comprising a plurality of 3D assets. The immersive training system and method compares, in real time, the received measured response of the user to a corresponding predetermined user response value to the 3D asset and, based on the user response comparison and a generated user-customized dynamic decision matrix, modifies at least one of the plurality of 3D assets of the 3D rendering. A customized dynamic decision matrix is generated for each user based on, for example, a corresponding personal performance profile and a neural efficiency score.

Abstract: A method for operating a self-orchestrated system for extraction, analysis, and presentation of entity data involves extracting a web page to object-based storage including web page content, web page metadata and a globally unique identifier. The method extracts the web page metadata from the object-based storage. The method inputs the web page metadata to a queue. The method pulls web page content from a content store. The method receives RegEx from a model parameter store. The method parses the web page content using RegEx and web page metadata. The method passes web page metadata and extracted content from the web page and positions of extracted content to an advanced analysis function decider (AAF Decider) for analysis. The method streams web page metadata and extracted content from the web page and positions of the extracted content to a JSON file batch for flattening.

Abstract: A computer-implemented method for providing agent assisted transcriptions of user utterances. A user utterance is received in response to a prompt provided to the user at a remote client device. An automatic transcription is generated from the utterance using a language model based upon an application or context, and presented to a human agent. The agent reviews the transcription and may replace at least a portion of the transcription with a corrected transcription. As the agent inputs the corrected transcription, accelerants are presented to the user comprising suggested texted to be inputted. The accelerants may be determined based upon an agent input, an application or context of the transcription, the portion of the transcription being replaced, or any combination thereof. In some cases, the user provides textual input, to which the agent transcribes an intent associated with the input with the aid of one or more accelerants.

Abstract: This invention deals with the next generation improvements in recommendation systems. Retailers want to grow their business and increase sales. One embodiment displays recommendations for inside sales during calls to prospects via a CRM. Another embodiment improves genomic cross-sell by summing correlations between attributes. A third embodiment improves cross-channel personalization by linking personal information, preferably via a one-way hash, to a unique customer ID. A fourth embodiment enables a common core mobile app for different retailers. A fifth embodiment identifies a shopper before purchase to provide personal recommendations while shopping. A sixth embodiment utilizes a market place with shared customers for customer acquisition. A seventh embodiment utilizes customers' preferences and characteristics and sales data to influence recommendations. The characteristics can be combined into a shopper psychographic persona to generate recommendations.

Abstract: This invention deals with a computer-implemented method for using predictive analytics to find a second set of entries in a first array that are related to a first set of entries in the first array without directly calculating entry-to-entry similarities between entries in the first array, the method including: (a) identifying a target entry in a second array, wherein the first set of entries in the first array are linked to the target entry in the second array; (b) finding likely entries in the second array that have similarities to a target entry in the second array using a correlation; (c) finding linked entries in the first array that are linked to the likely entries found in the second array; and (d) using the linked entries found in the first array as the second set of entries in the first array.

Abstract: Systems and methods are disclosed for determining a move driven by an interaction. In some embodiments, a processor determines an operational state of an interaction with a user based on parameter values of a data structure. The processor identifies a plurality of candidate moves for changing the operational state by determining a domain in which the interaction is occurring, retrieving a set of candidate moves that correspond to the domain from a knowledge graph, and adding the set to the plurality of candidate moves. The processor encodes input of the user received during the interaction into encoded terms, and determines a move for changing the operational state based on a match of the encoded terms to the set of candidate moves. The processor updates the parameter values of the data structure based on the move to reflect a current operational state led to by the move.

Abstract: A time-sensitive, data reduction system is described. According to an aspect of the system, a method is disclosed that comprises: receiving at least a portion of an airline-specific dataset, wherein the airline-specific dataset comprises a plurality of location records, wherein each location record comprises a point of interest (POI) and a unique rank, wherein the POI is defined by unique geographic identifier; determining a current position of the aircraft; and based on the current position and based on the unique rank of each of the plurality of location records, controlling, at a passenger display in the aircraft, a presentation of a map and a plurality of POI indicators associated with a highly-ranked subset of the plurality of location records. According to another aspect of the system, a plurality of servers are used provide relevant location-related content to an aircraft passenger.

Abstract: A masking system prevents a human agent from receiving sensitive personal information (SPI) provided by a caller during caller-agent communication. The masking system includes components for detecting the SPI, including automated speech recognition and natural language processing systems. When the caller communicates with the agent, e.g., via a phone call, the masking system processes the incoming caller audio. When the masking system detects SPI in the caller audio stream or when the masking system determines a high likelihood that incoming caller audio will include SPI, the caller audio is masked such that it cannot be heard by the agent. The masking system collects the SPI from the caller audio and sends it to the organization associated with the agent for processing the caller's request or transaction without giving the agent access to caller SPI.