Abstract: Described herein are techniques for using a graph neural network to encode online job postings as embeddings. First, an input graph is defined by processing one or more rules to discover edges that connect nodes in an input graph, where the nodes of the input graph represent job postings or standardized job attributes, and the edges are determined based on analyzing a log of user activity directed to online job postings. Next, a graph neural network (GNN) is trained based on an edge prediction task. Finally, once trained, the GNN is used to derive node embeddings for the nodes (e.g., job postings) of the input graph, and in some instances, new online job postings not represented in the original input graph.

Abstract: Computer-implemented methods, computer program products, and computer systems for mitigating frequency loss may include one or more processors configured for receiving first audio data corresponding to unobstructed user utterances, receiving second audio data corresponding to first obstructed user utterances, generating a frequency loss (FL) model representing frequency loss between the first audio data and the second audio data, receiving third audio data corresponding to one or more second obstructed user utterances, processing the third audio data using the FL model to generate fourth audio data corresponding to a frequency loss mitigated version of the second obstructed user utterances, and transmitting the fourth audio data to a recipient computing device. The first obstructed user utterances are obstructed by a facemask and the one or more second obstructed user utterances is obstructed by the facemask. The FL model may be executed as an audio plugin in a web conferencing program.

Abstract: Data processing apparatus comprises a data memory; a selection controller comprising a computer processor; and a digital interface between a control process implemented by the selection controller and a text handling process implemented by the computer processor or another processor; in which: the selection controller is configured to provide a text document from the data memory to the text handling process to identify one or more characteristics of words in the text document; the selection controller is configured to provide user selection of one or more of the words in the text document to be substituted and of one or more target characteristics; and the selection controller is configured to request from the text handling process a set of one or more substitute words for the selected words such that the substitute words comply with the selected one or more of the target characteristics.

Abstract: A method, computer program, and computer system to recover a dropped pronoun is provided for receiving data corresponding to one or more input words and determining contextual representations for the received input word data. The dropped pronoun may be identified based on a probability value associated with the contextual representations, and a span associated with one or more of the received input words may and that corresponds to which of the input words the dropped pronoun refers may be determined.

Abstract: An approach for generating synthetic treebanks to be used in training a parser in a production system is provided. A processor receives a request to generate one or more synthetic treebanks from a production system, wherein the request indicates a language for the one or more synthetic treebanks. A processor retrieves at least one corpus of text in which the requested language is present. A processor provides the at least one corpus to a transformer enhanced parser neural network model. A processor generates at least one synthetic treebank associated with a string of text from the at least one corpus of text in which the requested language is present. A processor sends the at least one synthetic treebank to the production system, wherein the production system trains a parser utilized by the production system with the at least one synthetic treebank.

Abstract: Implementations disclosed herein are directed to federated learning of machine learning (“ML”) model(s) based on gradient(s) generated at corresponding client devices and a remote system. Processor(s) of the corresponding client devices can process client data generated locally at the corresponding client devices using corresponding on-device ML model(s) to generate corresponding predicted outputs, generate corresponding client gradients based on the corresponding predicted outputs, and transmit the corresponding client gradients to the remote system. Processor(s) of the remote system can process remote data obtained from remote database(s) using global ML model(s) to generate additional corresponding predicted outputs, generate corresponding remote gradients based on the additional corresponding predicted outputs. Further, the remote system can utilize the corresponding client gradients and the corresponding remote gradients to update the global ML model(s) or weights thereof.

Abstract: This disclosure relates to a system and method for quantitative measure of subject specific sentiment analysis of a text input. The text input comprises subjects and objects. The text input is tokenized, and each word of the tokenized text input is tagged based on a part-of-speech (POS) and a universal dependency tag. A universal dependency tag tree is prepared based on dependency tags. Further, the subjects and objects are identified using a subject-verb-object (SVO) detection. The universal dependency tree is analyzed for each identified subject to determine a token dependency of the subject. The identified subject is quantified using a deep learning-based sentiment analyzer and finally a sentiment score is recommended for the subject using a probability score and a class score is assigned to the subject.

Abstract: A de-coupled computing infrastructure is described that is adapted to provide domain specific contextual engines based on conversational flow. The computing infrastructure further includes, in some embodiments, a mechanism for directing conversational flow in respect of a backend natural language processing engine. The computing infrastructure is adapted to control or manage conversational flows using a plurality of natural language processing agents.

Abstract: A system retrains a natural language understanding (NLU) model by regularly analyzing electronic documents including web publications such as online newspapers, blogs, social media posts, etc. to understand how word and phrase usage is evolving. Generally, the system determines the frequency of words and phrases in the electronic documents and updates an NLU dictionary depending on whether certain words or phrases are being used more frequently or less frequently. This dictionary is then used to retrain the NLU model, which is then applied to predict the meaning of text or speech communicated by a people group. By analyzing electronic documents such as web publications, the system is able to stay up-to-date on the vocabulary of the people group and make correct predictions as the vocabulary changes (e.g., due to natural disaster). In this manner, the safety of the people is improved.

Abstract: An interaction system that interacts with a user is disclosed. The interaction system includes: an input device that receives a speech signal of the user; a computing device that determines a speech content of the interaction system for a speech content acquired from the speech signal of the user such that a frequency distribution of speech feature values of the speech content of the interaction system approaches an ideal frequency distribution; and an output device that outputs the determined speech content of the interaction system.

Abstract: Disclosed are an emotion classification information-based text-to-speech (TTS) method and device. The emotion classification information-based TTS method according to an embodiment of the present invention may, when emotion classification information is set in a received message, transmit metadata corresponding to the set emotion classification information to a speech synthesis engine and, when no emotion classification information is set in the received message, generate new emotion classification information through semantic analysis and context analysis of sentences in the received message and transmit the metadata to the speech synthesis engine. The speech synthesis engine may perform speech synthesis by carrying emotion classification information based on the transmitted metadata.

Abstract: A named entity recognition method, a named entity recognition equipment and a medium are disclosed, the method including: acquiring a voice signal; extracting a voice feature vector in the voice signal; extracting, based on a literalness result after voice recognition is performed on the voice signal, a literalness feature vector in the literalness result; splicing the voice feature vector and the literalness feature vector to obtain a composite feature vector of each word in the voice signal; processing the composite feature vector of each word in the voice signal through a deep learning model to obtain a named entity recognition result.

Abstract: The present disclosure discloses a method performed at a terminal for reconstructing a speech signal, and a computer storage medium, and relates to the field of speech recognition. The method includes: collecting, by the terminal, a plurality of sound signals through a plurality of sensors of a microphone array; determining, by the terminal, a first speech signal in the plurality of sound signals; performing, by the terminal, signal separation on the first speech signal to obtain a second speech signal; and performing, by the terminal, reconstruction on the second speech signal through a distortion recovery model to obtain a reconstructed speech signal; the distortion recovery model being obtained by training based on a clean speech signal and a distorted speech signal. The embodiments of the present disclosure improve accuracy of speech recognition results.

Abstract: A method, computer program, and computer system to recover a dropped pronoun is provided for receiving data corresponding to one or more input words and determining contextual representations for the received input word data. The dropped pronoun may be identified based on a probability value associated with the contextual representations, and a span associated with one or more of the received input words may and that corresponds to which of the input words the dropped pronoun refers may be determined.