Abstract: A computer-implemented method includes converting tabular data to a text representation, generating metadata associated with the text representation of the tabular data, outputting one or more natural language data descriptions indicative of the tabular data in response to utilizing a large language model (LLM) and zero-shot prompting of the metadata and text representation of the tabular data, outputting one or more summaries utilizing the LLM and appending a prompt on the one or more natural language data descriptions, selecting a single summary of the one or more summaries in response to the single summary having a smallest validation rate, receiving a query associated with the tabular data, outputting one or more predictions associated with the query, and in response to meeting a convergence threshold with the one or more predictions generated from the one or more iterations, output a final prediction associated with the query.

Abstract: Computing systems of a multi-tenant trusted domain collect metadata describing data stored in data sources of a set of tenant trusted domains. The computing systems of the multi-tenant trusted domain use the metadata to process natural language questions based on data stored in data sources of a tenant trusted domain. The computing systems of the multi-tenant trusted domain identify a set of data sources of the tenant trusted domain that are relevant for processing the natural language question and generate an execution plan for answering the natural language question. The computing systems of the multi-tenant trusted domain send the execution plan to one or more computing systems of the tenant trusted domain. The computing systems of the tenant trusted domain execute the execution plan and send the result of executing the execution plan to a client device that sent the natural language question.

Abstract: Provided is a signal processing apparatus that includes a voice quality conversion unit that converts acoustic data of any sound of an input sound source to acoustic data of voice quality of a target sound source different from the input sound source on the basis of a voice quality converter parameter obtained by training using acoustic data for each of one or more sound sources as training data, the acoustic data being different from parallel data or clean data.

Abstract: Systems, methods, and computer-readable media are disclosed for dynamic voice search transitioning. Example methods may include receiving, by a computer system in communication with a display, a first incoming voice data indication, initiating a first user interface theme for presentation at a display, wherein the first user interface theme is a default user interface theme, and receiving first voice data. Example methods may include sending the first voice data to a remote server for processing, receiving an indication from the remote server to initiate a second user interface theme, and initiating the second user interface theme for presentation at the display.

Abstract: In non-limiting examples of the present disclosure, systems, methods and devices for training machine learning models are presented. An automated task framework comprising a plurality of machine learning models for executing a task may be maintained. A natural language input may be processed by two or more of the machine learning models. An action corresponding to a task intent identified from the natural language input may be executed. User feedback related to the execution may be received. The feedback may be processed by a user sentiment engine. A determination may be made by the user sentiment engine that a machine learning model generated an incorrect output. The machine learning model that generated the incorrect output may be identified. The machine learning model that generated the incorrect output may be automatically penalized via training. Any machine learning models that a user expressed neutral or positive sentiment toward may be rewarded.

Abstract: A method and device for signaling multiple audio mixing gains in a teleconference using a single Real-time Transport Protocol (RTP) header extension is provided. The method includes receiving an input audio stream from a 360-degree video/audio stream that includes multiple mixing gains from the input audio stream and overlay audio stream, declaring a single RTP header extension (including one or more of an element identifier, a length of an extension element, and a magnitude of the mixing gains) for all the mixing gains, and signaling the mixing gains using the single RTP header extension. The single RTP header extension may use a one-byte or two-byte header extension format which is declared using a Session Description Protocol (SDP).

Abstract: The disclosed technology is generally directed to the conversion of text to tables. In one example of the technology, input text that includes at least three rows is received. A plurality of characteristics of the input text is determined. Each characteristic of the plurality of characteristics is associated with a uniformity between the rows of the input text. The plurality of characteristics includes at least one characteristic that is associated with a delimiter count. A determination is made as to whether the input text is suitable for conversion to table based on the plurality of characteristics. Upon determining that the input text is suitable for conversion to a table, a machine learning model is used to convert the input text into a table.

Abstract: A method for decoding an encoded audio bitstream in an audio processing system is disclosed. The method includes extracting from the encoded audio bitstream a first waveform-coded signal comprising spectral coefficients corresponding to frequencies up to a first cross-over frequency for a time frame and performing parametric decoding at a second cross-over frequency for the time frame to generate a reconstructed signal. The second cross-over frequency is above the first cross-over frequency and the parametric decoding uses reconstruction parameters derived from the encoded audio bitstream to generate the reconstructed signal. The method also includes extracting from the encoded audio bitstream a second waveform-coded signal comprising spectral coefficients corresponding to a subset of frequencies above the first cross-over frequency for the time frame and interleaving the second waveform-coded signal with the reconstructed signal to produce an interleaved signal for the time frame.

Abstract: Systems and methods for deep technology innovation management by cross-pollinating innovations dataset are disclosed. A system extracts context-based keyword from an innovation dataset by transforming the innovation dataset to a vector. Further, the system searches semantically relevant keywords for the extracted context-based keyword, by extracting an entity and a key phrase from the extracted a context-based keyword. Furthermore, system clusters the vector, by identifying frequent keywords in the semantically relevant keywords to obtain cluster centroids of the frequent keywords. Thereafter, the system determines weighted keywords in each cluster using the obtained cluster centroids, and classifies the weighted keywords to identify emerging innovation trends relevant to the innovation in the innovation dataset. The system forms cohorts of innovators to explore the reuse of innovations, assets, code, and build focused monetization model.

Abstract: Improvement is made in performance of a trained neural network that uses positional information indicating a position at which each token included in an input sequence is present in the input sequence.

Abstract: Described herein is an Autonomous Conversational AI system, which does not require any human configuration or annotation, and is used to have multi-turn dialogs with a user. A typical Conversational AI system consists of three main models: Natural Language Understanding (NLU), Dialog Manager (DM) and Natural Language Generation (NLG), which requires human provided data and configuration. The system proposed herein leverages novel Conversational AI methods which automatically generates conversational AI configuration from any historical conversation logs. The automatically generated configuration contains Auto-Topics, Auto-Subtopics, Auto-Intents, Auto-Responses and Auto-Flows which are used to automatically train NLU, DM and NLG models. Once these models are trained for given conversation logs, the system can be used to have dialog with any user.

Abstract: Systems and methods are described herein for an application and graphical user interface (“GUI”) for customized storytelling. In an example, a user can create profiles for a listener user and a reader user. The listener user profile can include information about the listener user. The reader user profile can include a voice model of the reader user's voice. The GUI can allow the user to provide a brief description of a story. The application can send the story description and listener user profile to a server that uses an artificial intelligence engine to generate a customized story for the listener user. The application can apply the reader user voice model to the story and play audio of the reader user's voice reading the story.

Abstract: Implementations relate to generating a proficiency measure, and utilizing the proficiency measure to adapt one or more automated assistant functionalities. The generated proficiency measure is for a particular class of automated assistant actions, and is specific to an assistant device and/or is specific to a particular user. A generated proficiency measure for a class can reflect a degree of proficiency, of a user and/or of an assistant device, for that class. Various automated assistant functionalities can be adapted, for a particular class, responsive to determining the proficiency measure satisfies a threshold, or fails to satisfy the threshold (or an alternate threshold). The adaptation(s) can make automated assistant processing more efficient and/or improve (e.g., shorten the duration of) user-assistant interaction(s).

Abstract: Apparatuses, systems, and techniques to parse textual data using parallel computing devices. In at least one embodiment, text is parsed by a plurality of parallel processing units using a finite state machine and logical stack to convert the text to a tree data structure. Data is extracted from the tree by the plurality of parallel processors and stored.

Abstract: A method for neural network training is provided. The method inputs a training set of textual claims, lists of evidence including gold evidence chains, and claim labels labelling the evidence with respect to the textual claims. The claim labels include refutes, supports, and not enough information (NEI). The method computes an initial set of document retrievals for each of the textual claims. The method also includes computing an initial set of page element retrievals including sentence retrievals from the initial set of document retrievals for each of the textual claims. The method creates, from the training set of textual claims, a Leave Out Training Set which includes input texts and target texts relating to the labels. The method trains a sequence-to-sequence neural network to generate new target texts from new input texts using the Leave Out Training Set.

Abstract: Text editing apparatus comprises a database memory configured to store a text database, in which the text database is configured to store a plurality of text portions and a set of links between text portions, the set of links defining a document as a linked list of the text portions; and a data processor configured, in response to user input, to perform an editing operation to edit the text database so as to define an edited document by changing at least one of: (i) text within a text portion and (ii) the set of links between text portions.

Abstract: In certain embodiments, speech is converted to text for theme identification by natural language processing. Notification data is generated based on detected themes and the notification data may include rules for notification presentation on a client device. The notification data may include parameters for processing image data captured by an augmented reality device to detect one or more objects. The objects may be associated with the theme and detection thereof within captured image data, and in accordance with other rules, may cause the augmented reality device to present a notification with contextual relevance to a current environment of a user utilizing the augmented reality device.

Abstract: A system, method and computer product for training a neural network system. The method comprises applying an audio signal to the neural network system, the audio signal including a vocal component and a non-vocal component. The method also comprises comparing an output of the neural network system to a target signal, and adjusting at least one parameter of the neural network system to reduce a result of the comparing, for training the neural network system to estimate one of the vocal component and the non-vocal component. In one example embodiment, the system comprises a U-Net architecture. After training, the system can estimate vocal or instrumental components of an audio signal, depending on which type of component the system is trained to estimate.

Abstract: A system, method and computer product for training a neural network system. The method comprises inputting an audio signal to the system to generate plural outputs f(X, ?). The audio signal includes one or more of vocal content and/or musical instrument content, and each output f(X, ?) corresponds to a respective one of the different content types. The method also comprises comparing individual outputs f(X, ?) of the neural network system to corresponding target signals. For each compared output f(X, ?), at least one parameter of the system is adjusted to reduce a result of the comparing performed for the output f(X, ?), to train the system to estimate the different content types. In one example embodiment, the system comprises a U-Net architecture. After training, the system can estimate various different types of vocal and/or instrument components of an audio signal, depending on which type of component(s) the system is trained to estimate.

Abstract: In accordance with one embodiment, a system includes a processor, a memory module communicatively coupled to the processor, an NLP module communicatively coupled to the processor, and a set of machine-readable instructions stored in the memory module. The machine-readable instructions, when executed by the processor, direct the processor to perform operations including receiving a text data, and receiving a training text data for training one or more models of the NLP module. The operations also include generating, with a novice model of the NLP module, a novice suggestion based on the text data and the training text data to present an idea related to the text data, generating, with an expert model of the NLP module, an expert suggestion based on the text data and the training text data to present an idea elaborating on the text data, and outputting the novice suggestion and/or the expert suggestion.