Abstract: A system for theming is disclosed. A transformer receives a plurality of sampled thought objects and a prompt from a theming computer. The prompt is designed to provide context data, format requirements, and instructions for theme assignment. The plurality of thought objects comprises text present in data from communication environments and the prompt requests the transformer to assign themes to each thought object. An object-theming transformer with a list of themes determines the probability score for mapping a thought object to a list of themes. When an associated probability score is above a threshold, the thought object is assigned to a known theme. In an embodiment, un-themed thought objects (not mapped to known themes) are provided to a topic identification transformer to generate theme names. The themes assigned and/or generated by the theming transformer and topic identification transformer are displayed on a graphical user interface.

Abstract: An electronic device includes a memory configured to store computer-executable instructions; and a processor configured to execute the computer-executable instructions to: based on a result of analyzing a transmission environment of a wireless communication channel through which an audio signal is transmitted, determine a bitrate of the audio signal, encode the audio signal into packets according to the bitrate, the packets including a main packet for audio streaming and a plurality of extension packets for sound quality improvement, based on at least one of a type of the packets and the result of analyzing the transmission environment, determine a packet type indicating a modulation scheme and number of time slots used for transmitting each packet of the packets, and configure and transmit audio packets reflecting the packet type for each packet of the packets.

Abstract: A method of processing an acoustic signal is disclosed. According to one or more embodiments, a first acoustic signal is received via a first microphone. The first acoustic signal is associated with a first speech of a user of a wearable headgear unit. A first sensor input is received via a sensor, a control parameter is determined based on the sensor input. The control parameter is applied to one or more of the first acoustic signal, the wearable headgear unit, and the first microphone. Determining the control parameter comprises determining, based on the first sensor input, a relationship between the first speech and the first acoustic signal.

Abstract: Techniques for improving a semantic parser of a dialog system, by breaking the semantic parser into a coarse semantic parser and a fine semantic parser, are described. A method described herein includes accessing an utterance received in a dialog system. The utterance is a text-based natural language expression. The method further includes applying a coarse semantic parser to the utterance to determine an intermediate logical form for the utterance. The intermediate logical form indicates one or more intents in the utterance. The method further includes applying a fine semantic parser to the intermediate logical form to determine a logical form for the utterance. The logical form is a syntactic expression of the utterance according to an established grammar, and the logical form includes one or more parameters of the one or more intents. The logical form can be used to conduct a dialog with a user of the dialog system.

Abstract: Technology is provided for causing a computing system to extract conversation features from a multiparty conversation (e.g., between a coach and mentee), apply the conversation features to a machine learning system to generate conversation analysis indicators, and apply a mapping of conversation analysis indicators to actions and inferences to determine actions to take or inferences to make for the multiparty conversation. In various implementations, the actions and inferences can include determining scores for the multiparty conversation such as a score for progress toward a coaching goal, instant scores for various points throughout the conversation, conversation impact score, ownership scores, etc. These scores can be, e.g., surfaced in various user interfaces along with context and benchmark indicators, used to select resources for the coach or mentee, used to update coach/mentee matchings, used to provide real-time alerts to signify how the conversation is going, etc.

Abstract: A system and method for updating computerized language models is provided that automatically adds or deletes terms from the language model to capture trending events or products, while maximizing computer efficiencies by deleting terms that are no longer trending and use of knowledge bases, machine learning model training and evaluation corpora, analysis tools and databases.

Abstract: Systems and methods are disclosed comprising techniques for signal processing, such as determining domain groups for portions of a signal and applying domain-specific signal quality control rules to various portions of a signal. The techniques can include receiving audio signal data corresponding to a recorded interaction, converting the audio signal data into a transcript that includes alphanumeric components, prompting a generative machine learning model to generate a response that maps at least one alphanumeric component of the converted transcript to a target signal domain group, prompting a generative machine learning model to generate a response that includes a set of alphanumeric elements from the at least one alphanumeric component that satisfy the at least one signal extraction rule of the target signal domain group, and generating a computer-based prediction for a set of attributes for the at least one alphanumeric component.

Abstract: A system configured to perform neural sidelobe cancelling to achieve enhanced target speech separation. A device improves target speech separation by combining a fixed directional beamformer with a deep neural network (DNN) configured to (i) selectively enhance speech in one target direction, (ii) cancel speech from non-target directions, and (iii) cancel non-speech interference from all directions. By combining the directional output from the beamformer with raw microphone signals as input signals to the DNN, the DNN is given a directional cue with which to select the target speech. The DNN generates mask data representing the target speech using a combination of temporal convolutional blocks and causal self-attention blocks. For example, the DNN achieves low latency by processing a large number of short frames, while the causal self-attention blocks smooth the output and reinforce local consistency of speech.

Abstract: Systems and methods are provided to compare a target sample of text to a set of textual records, each textual record including a sample of text and an indication of one or more segments of text within the sample of text. Semantic similarity values between the target sample of text and each of the textual records are determined. Determining a particular semantic similarity value between the target sample of text and a particular textual record of the corpus includes: (i) determining individual semantic similarity values between the target sample of text and each of the segments of text indicated by the particular textual record, and (ii) generating the particular semantic similarity value between the target sample of text and the particular textual record based on the individual semantic similarity values. A textual record is then selected based on the semantic similarities.

Abstract: Technology is provided for identifying synthesized conversation features from recorded conversations. The technology can identify, for each of one or more utterances, data for multiple modalities, such as acoustic data, video data, and text data. The technology can extract features, for each particular utterance of the one or more utterances, from each of the data for the multiple modalities associated with that particular utterance. The technology can also apply a machine learning model that receives the extracted features and/or previously synthesized conversation features and produces one or more additional synthesized conversation features.

Abstract: This disclosure relates to systems and methods for multi-utterance generation of data. Embodiments of the present disclosure utilizes a smart framework that is capable of generating contextually relevant utterances with immutability regulation and punctuation-memory. More Specifically, the present disclosure generates multiple syntactically and semantically correct utterances for text input data in such a way that a provision to selectively maintain or regulate phrases or words intact is provided and punctuation consistency is maintained.

Abstract: An apparatus processing an encoded signal to acquire first and second frames comprising spectral values and an aliasing portion; applying a transform to the first frame using a first window function to acquire a first block of samples, applying another transform to a first portion of the second frame using a second window function, and applying another one or more transforms to a second portion of the second frame using one or more third window functions to acquire a second block of samples; and post-processing the second block of samples using a folding-out operation to acquire a post-processed second block of samples comprising a portion of the second block of samples overlapping with the first block of samples in a multi-overlap region, windowing the post-processed second block of samples using an auxiliary window function, and overlap-adding a windowed post-processed second block of samples and the first block of samples.

Abstract: A method of risk score comparison and analysis using wearable devices is illustrated. The method comprising receiving, at a computing device, user data associated with a first user, generating, using the computing device, a first risk score as a function of the user data, and matching, using the computing device, the first user to a second user having a second risk score as a function of user data associated with the second user. Matching the users comprises extracting, from the first risk score, a plurality of first risk score content elements, extracting, from the second risk score, a plurality of second risk score content elements, and matching the plurality of first risk score elements to the plurality of second risk score elements.

Abstract: Systems and methods for creating predictor variables from unstructured data for prediction models are provided. A variable creation application receives unstructured data and processing the unstructured data to generate processed data. Based on the processed data, the variable creation application generates an attribute pool that contains multiple predictor variables generated by applying natural language processing (NLP) procedures on the processed data. The variable creation application further executes a prediction model on at least the predictor variables in the attribute pool to generate a prediction result. Based on the prediction result, the variable creation application evaluates the predictive power of each of the predictor variables and retains predictor variables that are predictive as input predictor variables for the prediction model.

Abstract: This disclosure relates to systems and methods for processing content and, particularly, but not exclusively, systems and methods for processing audio content. Systems and methods are described that provide techniques for processing, analyzing, and/or structuring of longer-form content to, among other things, make the content searchable, identify relevant and/or interesting segments within the content, provide for and/or otherwise generate search results and/or coherent shorter-form summaries and/or highlights, enable new shorter-form audio listening experiences, and/or the like. Various aspects of the disclosed systems and methods may further enable relatively efficient transcription and/or indexing of content libraries at scale, while also generating effective formats for users interacting with such libraries to engage with search results.

Abstract: A home appliance including a first microphone that is disposed on a surface of a housing, a second microphone that is disposed on an inside of the housing, and a processor configured to perform signal processing for first voice data that is acquired from the first microphone, and perform voice recognition using the signal-processed first voice data. The processor is further configured to generate noise data using second voice data that is acquired from the second microphone and perform signal processing for the first voice data using the generated noise data.

Abstract: A device includes a memory configured to store instructions and one or more processors configured to execute the instructions. The one or more processors are configured to execute the instructions to obtain link data corresponding to a communication link to a second device. The one or more processors are configured to execute the instructions to select, at least partially based on the link data, between an ambisonics mode and a stereo mode.

Abstract: Methods, apparatus and articles of manufacture to identify sources of network streaming services are disclosed. An example method includes receiving a first audio signal that represents a decompressed second audio signal, identifying, from the first audio signal, a parameter of an audio compression configuration used to form the decompressed second audio signal, and identifying a source of the decompressed second audio signal based on the identified audio compression configuration.

Abstract: A speech instruction recognition method, an electronic device, and a non-transient computer readable storage medium. The speech instruction recognition method comprises: acquiring a target speech; processing the target speech to obtain a target speech vector corresponding to the target speech; performing speech recognition on the target speech to obtain a target speech text of the target speech, and processing the target speech text to obtain a target text vector corresponding to the target speech text; and inputting the target speech vector and the target text vector to a pre-trained instruction recognition model to obtain an instruction category corresponding to the target speech.

Abstract: Methods, apparatus and articles of manufacture to identify sources of network streaming services are disclosed. An example apparatus includes a coding format identifier to identify, from a received first audio signal representing a decompressed second audio signal, an audio compression configuration used to compress a third audio signal to form the second audio signal, and a source identifier to identify a source of the second audio signal based on the identified audio compression configuration.