Samsung Electronics Co., Ltd.

Jehun Jeon

GyuBum Han

Jeong-Hoon Park

Young-Seok Kim

Junhwi Choi

Sang Hyun Yoo

Inkyu Choi

Jung-Hoe Kim

Jangsu Lee

Misuk Kim

Kyungtae Kim

Jaeyung Yeo

Hoshik Lee

Kyoung Gu Woo

Hyunjin Park

Sanghyun Yoo

Kyounggu Woo

Hojun Jaygarl

Seongmin Ok

Sangwook Kang

Jiseung Jeong

Disclosed is an entity linking method. A method includes: extracting an entity from an input context including text stored in a memory; obtaining candidate entities corresponding to, and based on, the extracted entity; determining a keyword based on the input context; generating keyword-based entity information based on the keyword and based on the extracted entity; and determining a top-matching entity corresponding to the entity based on the keyword-based entity information and the plurality of candidate entities.

Method and apparatus with entity linking

A method and apparatus for training an embedding vector generation model are provided, the method includes identifying a keyword in a query sentence, generating an embedding vector of the query sentence and an embedding vector of the keyword based on the embedding vector generation model, and training the embedding vector generation model such that a first similarity between the embedding vector of the query sentence and the embedding vector of the keyword is greater than a second similarity between an embedding vector of a reference sentence that does not include the keyword and the embedding vector of the keyword.

Method and apparatus for training embedding vector generation model

A user preference modeling method including receiving preference scores corresponding to items, receiving an input for selecting an item from among the items, decaying a preference score corresponding to one or more items from among the items included in a first list and a second list based on a time decay rate and a first parameter, in response to the selected item being included in the first list, decaying the preference score corresponding to the one or more items comp included in the first list and the second list based on a time decay rate and a second parameter, in response to the selected item being included in the second list, and increasing a preference score corresponding to the selected item, wherein the first list may include one or more of the plurality of items based on the preference scores, and wherein the first list is different from the second list.

Method and apparatus for modeling user preferences

A method and device with natural language processing is disclosed. The method includes performing a word embedding of an input sentence, encoding a result of the word embedding, using an encoder of a natural language processing model, to generate a context embedding vector, decoding the context embedding vector, using a decoder of the natural language processing model, to generate an output sentence corresponding to the input sentence, generating a score indicating a relationship between the context embedding vector and each of a plurality of knowledge embedding vectors, determining a first loss based on the output sentence, determining a second loss based on the generated score, and performing training of the natural language processing model, including training the natural language processing model based on the determined first loss, and training the natural language processing model based on the determined second loss.

Method and device with natural language processing

Disclosed is an operation method of a neural network including a first network and a second network, the method including acquiring state information output from the first network based on input information, determining whether the state information satisfies a condition using the second network, iteratively applying the state information to the first network in response to determining that the state information does not satisfy the condition, and outputting the state information in response to determining that the state information satisfy the condition.

Operating method and training method of neural network and neural network thereof

A speech generation method and apparatus are disclosed. The speech generation method includes obtaining, by a processor, a linguistic feature and a prosodic feature from an input text, determining, by the processor, a first candidate speech element through a cost calculation and a Viterbi search based on the linguistic feature and the prosodic feature, generating, at a speech element generator implemented at the processor, a second candidate speech element based on the linguistic feature or the prosodic feature and the first candidate speech element, and outputting, by the processor, an output speech by concatenating the second candidate speech element and a speech sequence determined through the Viterbi search.

Method and apparatus for generating speech

A method and apparatus with natural language processing is disclosed. The method includes determining a first similarity between an input sentence of a user and a select first database query sentence and dependent on a determination that the first similarity fails to meet a first threshold, determining a second similarity between a portion of the input sentence, less than all of the input sentence, and a select second database query sentence, and in response to the second similarity meeting a second threshold, outputting a response sentence corresponding to the second database query sentence as a response to the input sentence.

Method and apparatus with natural language processing

An electronic device and an method of the electronic device are provided, where the electronic device maintains a context that does not reflect a request for a secret conversation, in response to the request for the secret conversation being received from a first user, and generates a response signal to a voice signal of a second user based on the maintained context, in response to an end of the secret conversation with the first user.

Electronic device and method for conducting dialog

The present invention can receive a user input comprising a request to perform a task using at least one of the application programs via at least one of a display or a microphone, transmit data related to the user input to an external server through a communication circuit, transmit version information on at least one of the application programs to the external server via the communication circuit together with the data transmission or after the data transmission, receive a pass rule as a response to the request from the external server via the communication circuit, and store instructions that cause an electronic device to perform the task by allowing the electronic device to have the sequence. The pass rule comprises information on a sequence of states of the electronic device for performing the task, and at least one of the states may be one determined by the external server on the basis of at least a portion of the version information.

Method and apparatus for managing intelligent agent

A processor-implemented response inference method and apparatus are disclosed. The response inference apparatus receives an input, generates a latent variable vector in a latent variable region space by encoding the input, generates a validation vector with a predetermined phase difference from the latent variable vector, generates an output response by decoding the latent variable vector, generates a validation response by decoding the validation vector, and validates the output response by comparing the output response to the validation response.

Response inference method and apparatus

Provided is a processor-implemented method of generating a natural language, the method including generating a latent variable from an embedding vector that corresponds to an input utterance, determining attention information related to the input utterance by applying the generated latent variable to a neural network model, and outputting a natural language response that corresponds to the input utterance based on the calculated attention information.

Method and apparatus with natural language generation

An out-of-service (OOS) sentence generating method includes: training models based on a target utterance template of a target service and a target sentence generated from the target utterance template; generating a similar utterance template that is similar to the target utterance template based on a trained model, among the trained models, and a sentence generated from an utterance template of another service; and generating a similar sentence that is similar to the target sentence based on another trained model, among the trained models, and the similar utterance template.

OOS sentence generating method and apparatus

A voice recognition apparatus and corresponding method include a processor configured to calculate a probability distribution corresponding to an intent associated with an utterance of a user by applying pre-stored training data to an input voice signal input based on the utterance. The processor is also configured to select a target feature extractor including either one or both of a training-based feature extractor and a rule-based feature extractor using the calculated probability distribution, and extract a feature associated with the utterance based on the selected target feature extractor.

Voice recognition apparatus and method

An intention analysis apparatus includes an extractor configured to extract a feature value from a text input corresponding to machine recognition of an audio signal, a verifier configured to verify at least one state value associated with at least one of the text input or the audio signal, and a trained calculator configured to calculate a probability distribution of a user intention corresponding to the audio signal based on the feature value and the at least one state value.

Growing community of inventors

Suwon-si, South Korea

Jehun Jeon