Industry-University Cooperation Foundation Hanyang University

Jae Uk Lee

Seung Min Lee

Jung Hwan Kim

Jung Sik Kim

Jinho Ahn

Joon Hyuk Chang

Hyun Min Song

Seongchul Hong

According to an aspect, method for synthesizing multi-speaker speech using an artificial neural network comprises generating and storing a speech learning model for a plurality of users by subjecting a synthetic artificial neural network of a speech synthesis model to learning, based on speech data of the plurality of users, generating speaker vectors for a new user who has not been learned and the plurality of users who have already been learned by using a speaker recognition model, determining a speaker vector having the most similar relationship with the speaker vector of the new user according to preset criteria out of the speaker vectors of the plurality of users who have already been learned, and generating and learning a speaker embedding of the new user by subjecting the synthetic artificial neural network of the speech synthesis model to learning, by using a value of a speaker embedding of a user for the determined speaker vector as an initial value and based on speaker data of the new user.

Method and apparatus for synthesizing multi-speaker speech using artificial neural network

A mask for an extreme ultraviolet (EUV) lithography process is provided. The mask includes a substrate, a reflection layer including first material layers and second material layers which are alternately and repeatedly stacked on the substrate, a capping layer on the reflection layer, and a phase shift layer and an absorber layer sequentially stacked on the capping layer. Sidewalls of the phase shift layer and the absorber layer may be oblique to a top surface of the capping layer.

Mask for extreme ultraviolet lithography process and method of fabricating the same

Growing community of inventors

Yangsan-si, South Korea

Jae Uk Lee