Abstract: Provided is an electronic apparatus and a method of managing feature information performed by the electronic apparatus, the method including the electronic apparatus receiving a request to store feature information in a database, based on at least some of a feature quantity, attributes of the feature information, a calculation method of the feature information and sources of the feature information, the electronic apparatus storing the feature information in one of a first database in which data for inference of a machine learning model is stored and a second database in which data for training the machine learning model is stored, and the electronic apparatus synchronizing the feature information from one of the first database and the second database in which the feature information is stored to the other database.

Abstract: A training method of a conversation model according to various example embodiments of the present disclosure may include identifying a first context, identifying a first response set corresponding to the first context based on a first model, identifying a response subset selected from the first response set based on a gold response corresponding to the first context and training a second model based on the first context information and the response subset.

Abstract: The present invention discloses a method of providing information in a terminal of a user, comprising: a step in which an input is obtained requesting to use a matching service for matching with another person; a step in which it is checked whether to provide a live broadcast participation service to the user in response to the input requesting to use the matching service; and a step in which, if it is confirmed to provide the live broadcast participation service to the user, information is provided to the user related to participation in a live broadcast; and a step in which one or more items of content associated with the user are provided to a viewer of a live broadcast room in which the user participates.

Abstract: Disclosed is a method of training a dialogue model in an electronic device, the method including selecting a first context from a first dialogue data set including at least one pair of a context and a response corresponding to the context, generating a first response corresponding to the first context through a first dialogue model, generating an augmented dialogue dataset by incorporating a pair of the first context and the first response corresponding to the first context into the first dialogue data set, and training a second dialogue model based on the augmented dialogue dataset.

Abstract: Provided is an apparatus for evaluating and improving responses, and a method and a computer readable recording medium thereof. The apparatus for evaluating responses according to the present disclosure obtains cluster classifying information for training responses, and based on distribution of clusters to which test responses output from the dialogue generation model are classified, evaluate semantic diversity of the responses output from the dialogue generation model.

Abstract: Provided is an apparatus for evaluating and improving responses, and a method and a computer readable recording medium thereof. The apparatus for evaluating responses according to the present disclosure obtains cluster classifying information for training responses, and based on distribution of clusters to which test responses output from the dialogue generation model are classified, evaluate semantic diversity of the responses output from the dialogue generation model.

Abstract: Provided is an electronic apparatus and a method of managing feature information performed by the electronic apparatus, the method including the electronic apparatus receiving a request to store feature information in a database, based on at least some of a feature quantity, attributes of the feature information, a calculation method of the feature information and sources of the feature information, the electronic apparatus storing the feature information in one of a first database in which data for inference of a machine learning model is stored and a second database in which data for training the machine learning model is stored, and the electronic apparatus synchronizing the feature information from one of the first database and the second database in which the feature information is stored to the other database.

Abstract: Provided is a method of transforming a landmark including: receiving an input image including a facial image of a first person and a landmark corresponding to the facial image; estimating a transformation matrix corresponding to the landmark; and calculating an expression landmark and an identity landmark corresponding to the input image by using the transformation matrix.

Abstract: Systems and techniques to generate imitative responses are illustrated. response generation method performed in an electronic apparatus of the present disclosure includes acquiring at least one piece of utterance data, acquiring a first context corresponding to the utterance data from a context candidate set, generating one or more dialogue sets including the first context and the utterance data, receiving a second context from a user, and acquiring a response corresponding to the second context using a language model based on the one or more dialogue sets.

Abstract: A training method of a conversation model according to various example embodiments of the present disclosure may include identifying a first context, identifying a first response set corresponding to the first context based on a first model, identifying a response subset selected from the first response set based on a gold response corresponding to the first context and training a second model based on the first context information and the response subset.

Abstract: Provided is a method for providing information by a recommendation server, the method including obtaining first information regarding a user and second information regarding broadcasts from a service server regarding broadcasts, receiving a request by the user for a recommended broadcast list from the service server, identifying, in response to the request, third information about a live-streaming broadcast based on the second information, and providing recommended broadcast list information based on the first information and the third information.

Abstract: Disclosed is a method of training a dialogue model in an electronic device, the method including selecting a first context from a first dialogue data set including at least one pair of a context and a response corresponding to the context, generating a first response corresponding to the first context through a first dialogue model, generating an augmented dialogue dataset by incorporating a pair of the first context and the first response corresponding to the first context into the first dialogue data set, and training a second dialogue model based on the augmented dialogue dataset.

Abstract: A method of generating a reenacted image includes: extracting a landmark from each of a driver image and a target image; generating a driver feature map based on pose information and expression information of a first face shown in the driver image; generating a target feature map and a pose-normalized target feature map based on style information of a second face shown in the target image; generating a mixed feature map by using the driver feature map and the target feature map; and generating the reenacted image by using the mixed feature map and the pose-normalized target feature map.

Abstract: Disclosed is a speech synthesis method including acquiring second speech data and a target text, acquiring first information includes embedding information corresponding to the second speech data, acquiring second information including embedding information of the second speech data, the embedding information in relation with with components generated based on a sequence of the target text, and acquiring audio data corresponding to the target text and reflecting characteristics of speech of a speaker based on the first information and the second information.

Abstract: Provided is a method of transforming a landmark including: receiving an input image including a facial image of a first person and a landmark corresponding to the facial image; estimating a transformation matrix corresponding to the landmark; and calculating an expression landmark and an identity landmark corresponding to the input image by using the transformation matrix.

Abstract: Systems and techniques to generate imitative responses are illustrated. response generation method performed in an electronic apparatus of the present disclosure includes acquiring at least one piece of utterance data, acquiring a first context corresponding to the utterance data from a context candidate set, generating one or more dialogue sets including the first context and the utterance data, receiving a second context from a user, and acquiring a response corresponding to the second context using a language model based on the one or more dialogue sets.

Abstract: A method of generating a reenacted image includes: extracting a landmark from each of a driver image and a target image; generating a driver feature map based on pose information and expression information of a first face shown in the driver image; generating a target feature map and a pose-normalized target feature map based on style information of a second face shown in the target image; generating a mixed feature map by using the driver feature map and the target feature map; and generating the reenacted image by using the mixed feature map and the pose-normalized target feature map.

Abstract: Disclosed is a speech synthesis method including acquiring second speech data and a target text, acquiring first information includes embedding information corresponding to the second speech data, acquiring second information including embedding information of the second speech data, the embedding information in relation with components generated based on a sequence of the target text, and acquiring audio data corresponding to the target text and reflecting characteristics of speech of a speaker based on the first information and the second information.

Abstract: A report evaluation method of a report evaluation device may include receiving a report from at least one client terminal, determining a category of the received report, identifying a learning model corresponding to the category, evaluating a reliability of the report through the learning model, and generating and outputting information on the reliability.

Abstract: A terminal may include a display that is divided into at least two areas, when a real time broadcasting, where a user of the terminal is a host, starts through a broadcasting channel, and of which one area of the at least two areas is allocated to the host; an input/output interface that receives a voice of the host; a communication interface that receives one item selected of at least one or more items and a certain text from a terminal of a certain guest, of at least one or more guests who entered the broadcasting channel; and a processor that generates a voice message converted from the certain text into the voice of the host or a voice of the certain guest.