Abstract: Computing units provided at local sites or edge locations are programmed to execute conversational tools that generate pertinent, domain-specific responses to queries received from workers at such sites or locations. The conversational tools are large language models that are trained with domain-specific knowledge documents. Data representing queries are received from workers at such sites or locations and provided as inputs to the conversational tools along with text representing nearest knowledge documents from a knowledge base associated with the domain, as well as contextual data. Responses identified based on outputs received from the conversational tools in response to the inputs are provided to the workers that generated the queries.

Abstract: An apparatus for decoding data segments representing a time-domain data stream, a data segment being encoded in the time domain or in the frequency domain, a data segment being encoded in the frequency domain having successive blocks of data representing successive and overlapping blocks of time-domain data samples. The apparatus includes a time-domain decoder for decoding a data segment being encoded in the time domain and a processor for processing the data segment being encoded in the frequency domain and output data of the time-domain decoder to obtain overlapping time-domain data blocks. The apparatus further includes an overlap/add-combiner for combining the overlapping time-domain data blocks to obtain a decoded data segment of the time-domain data stream.

Abstract: An example process includes at an electronic device with one or more processors and memory: while the electronic device is in a locked state and in communication with an external accessory device: receiving a natural language input; determining whether the natural language input corresponds to a secure task request; in accordance with a determination that the natural language input corresponds to a secure task request: determining whether one or more states of the external accessory device satisfy a set of authentication criteria; and in accordance with a determination that the one or more states of the external accessory device satisfy the set of authentication criteria: initiating, by the electronic device, a secure task corresponding to the natural language input; and providing an output indicative of the initiated secure task.

Abstract: Provided is an information processing device comprising an estimation unit that estimates a phase regarding the user's decision making, and a response generation unit that generates a response to the user's utterance according to the estimated phase. The present technology can be applied to, for example, smart speakers and smart displays.

Abstract: To provide a voice analysis system capable of performing voice recognition with higher accuracy. A voice analysis system including a first voice analysis terminal and a second voice analysis terminal, the first voice analysis terminal obtaining first conversation information, and the second voice analysis terminal obtaining second conversation information, wherein the voice analysis system comprises a conversation category selection unit which compares the number of related words included in the first conversation information and the number of related words included in the second conversation information, in each conversation category, and adopts the conversation category with the larger number of related words as a correct conversation category.

Abstract: Certain aspects of the present disclosure provide techniques for orchestrating a user experience using natural language input. A user experience is orchestrated within an ecosystem of features for which a plurality of corresponding tokens is defined. Natural language describing a desired user experience result is received by a user experience orchestrator. A sequence of tokens corresponding to operations belonging to an ecosystem of features which produce a correct result for the natural language input can be identified by a trained large language model and executed by the user experience orchestrator using a token operator. The output operations determined by the model to produce or be likely to produce the correct result based on the natural language input can be disambiguated, confirmed, and/or executed.

Abstract: A method including receiving a user input for input to a language processing machine learning model (MLM). The method also includes generating modified inputs that are based on, and semantically related to, the user input. The method also includes executing the MLM to generate model outputs of the MLM. The MLM takes as input instances of each of the modified inputs. The method also includes sampling the model outputs using a statistical sampling strategy to generate sampled model outputs. The method also includes clustering the sampled model outputs into clusters. Each cluster of the clusters represents a distinct semantic meaning of the sampled model outputs. The method also includes generating a confidence metric for the user input. The confidence metric includes a predictive entropy of the clusters. The method also includes routing the user input based on whether the confidence metric satisfies or fails to satisfy a threshold value.

Abstract: According to an embodiment, an electronic device comprises: a memory, a communication module comprising communication circuitry, and a processor operatively connected with the memory and the communication module. The processor is configured to control the electronic device to: obtain a utterance text corresponding to utterance speech, obtain an intent of the utterance text and emotion information based on the utterance speech and the utterance text, obtain a response text for the utterance text based on the intent of the utterance text and the emotion information, obtain a markup language including information about an output unit of text of the response text based on at least one of the intent of the utterance text, the emotion information, or the response text, and add the markup language to the response text and provide the response text. The text output unit is at least one selected from among a phoneme unit, a consonant and vowel unit, a syllable unit, or a word unit.

Abstract: Aspects of the present disclosure relate to systems and methods for detecting emerging events. In various examples, a method for detecting emerging events includes obtaining communication data associated with communication between multiple sources, segmenting communication data into multiple segments, determining whether a data segment belongs to a familiar topic or none, and generating a notification when a familiar topic is mentioned for more or less than a mention prediction. Additionally, or alternatively, a notification may be generated when an unfamiliar topic emerges from a set of unfamiliar data segments if an associated segment count exceeds a critical mass threshold. To determine whether a data segment belongs to a familiar topic, the data segment may be transformed into a feature vector and mapped onto a feature space, where a distance-based similarity score may be determined.

Abstract: The disclosure provides a role labeling method. The method includes: obtaining a dialogue sentence to be labeled and context information corresponding to the dialogue sentence, and splicing the context information and the dialogue sentence to obtain a spliced text; extracting location information of a role name of the dialogue sentence in the spliced text from the spliced text; determining a first candidate role name of the dialogue sentence based on the location information; determining a second candidate role name of the dialogue sentence from role names in the spliced text; and determining a target role name of the dialogue sentence based on the first candidate role name and the second candidate role name, and performing role labeling on the dialogue sentence based on the target role name.

Abstract: In the present disclosure, provided are an electronic apparatus and a controlling method therefor. The electronic apparatus of the present disclosure comprises: a memory in which a first model and a second model are stored; and a processor configured to, when user voice is received, generate a response to the user voice according to completion of the received user voice, the completion being determined on the basis of the first model and the second model. Here, the first model uses, as training data, sentence structures of a plurality of completed sentences, and the second model uses, as training data, a plurality of completed sentences and intents corresponding to the plurality of completed sentences. In particular, the first model and the second model may be artificial intelligence models trained according to at least one of machine learning, a neural network, or a deep learning algorithm.

Abstract: An audio signal encoding method is provided comprising: receiving first and second audio signal frames; processing a second portion of the first audio signal frame and a first portion of the second audio signal frame using an orthogonal transformation to determine in part a first intermediate encoding result; and processing the first intermediate encoding result using an orthogonal transformation to determine a set of spectral coefficients that corresponds to at least a portion of the first audio signal frame.

Abstract: A method includes detecting, in a written electronic communication, an input sentence satisfying a readability metric threshold, and processing, by a sentence transformer model responsive to the input sentence satisfying the readability metric threshold, the input sentence to output a suggested set of sentences. The method further includes evaluating the first suggested set of sentences along a set of acceptability criteria, and determining, based on the evaluating, that the set of acceptability criteria is satisfied. The method further includes modifying, based on determining that the set of acceptability criteria is satisfied, the written electronic communication with the suggested set of sentences to obtain a modified written electronic communication, and storing the modified written electronic communication.

Abstract: The one or more embodiments provide for a method, system, and computer program product, an intent, generated by a large language model from a text, is received from a user device as a first input to an advice planner. A state of an account is received as a second input to the advice planner. The advice planner classifieds the intent into a domain corresponding to the intent, and generates, as output, a plan comprising a first set of action logic associated with the domain. Each action logic is a discrete step in an ordered sequence for achieving a desired state of the account. The advice planner forwards the plan to the large language model (LLM). The large language model receives the plan as input and generates advice in a natural language format as output. The advice is then forwarded to the user device.

Abstract: A virtual assistant system comprising an interface configured to receive user input and provide a response to the user and a processor configured to run machine executable code. A memory storing non-transitory machine executable code configured to process the user input to generate two or more primary interpretations and one or more secondary interpretations based on one or more of the two or more primary interpretations. The code is also configured to process the primary interpretations and alternative interpretations to generate results which lead to two or more terminal states and then score the two or more terminal states to rank the two or more terminal states such that a top ranked terminal state is the top result, which is presented to the user. A transceiver may communicate over a network to a second device configured to assist the virtual assistant system in generating the top result for the user.

Abstract: An information processing apparatus outputs answer information corresponding to inquiry information that is input. The information processing apparatus includes a memory and circuitry. The memory is configured to store a plurality of databases each having at least a first field and a second field. The circuitry is configured to: perform morphological analysis on the inquiry information, to divide the inquiry information into morphemes; perform a first matching process based on the morphemes and the first field of each of the plurality of databases, to determine whether to adopt the database as an extraction source from which the answer information is to be extracted; and perform a second matching process based on the morphemes and the first field of the database, which is determined to be adopted as the extraction source, to output, as the answer information, data in the second field corresponding to data in the first field.

Abstract: A query request identifying a query and a plurality of text portions for determining an answer to the query may be received. Relevance scores corresponding with respective ones of the text portions may be determined based on application of one or more machine learning models to the respective text portion and the query. A subset of the text portions may be selected based on the relevance scores. A response message to the query request including an answer to the query in natural language text generated by a large language model based on the first subset of text portions may be determined.

Abstract: The invention provides an audio encoder including a combination of a linear predictive coding filter having a plurality of linear predictive coding coefficients and a time-frequency converter, wherein the combination is configured to filter and to convert a frame of the audio signal into a frequency domain in order to output a spectrum based on the frame and on the linear predictive coding coefficients; a low frequency emphasizer configured to calculate a processed spectrum based on the spectrum, wherein spectral lines of the processed spectrum representing a lower frequency than a reference spectral line are emphasized; and a control device configured to control the calculation of the processed spectrum by the low frequency emphasizer depending on the linear predictive coding coefficients of the linear predictive coding filter.

Abstract: The present invention relates to a method for receiving data transmitted acoustically. The method includes receiving an acoustically transmitted signal encoding data; processing the received signal to minimise environmental interference within the received signal; and decoding the processed signal to extract the data. The data encoded within the signal using a sequence of tones. A method for encoding data for acoustic transmission is also disclosed. This method includes encoding data into an audio signal using a sequence of tones. The audio signal in this method is configured to minimise environmental interference. A system and software are also disclosed.

Abstract: A computer-implemented method, a computer program product, and a computer system for keyword extraction. A computer vectorizes word sets by splitting text in a corpus. The computer trains a Gaussian mixture distribution model for the word sets and obtains an inverse document frequency (IDF) value of a word set from the learned Gaussian mixture distribution model. The computer trains a Gaussian mixture distribution model for a cluster and obtains a term frequency (TF) value of the word set in the cluster from the learned Gaussian mixture distribution model. The computer calculates a term frequency-inverse document frequency (TF-IDF) value of the word set in the cluster, based on the TF value and the IDF value. The computer calculates TF-IDF values of the word sets in clusters and rearranges the word sets in a descending order of the TF-IDF values to obtain extracted keywords.