Open AccessSource Separation using Sparse NMF and Graph Regularization on Vietnamese Dataset
Author(s) -
Phạm Minh Tuấn
Publication year - 2020
Publication title -
tạp chí khoa học và công nghệ
Language(s) - English
Resource type - Journals
ISSN - 1859-1531
DOI - 10.31130/ict-ud.2020.98
Subject(s) - vietnamese , non negative matrix factorization , regularization (linguistics) , computer science , source separation , artificial intelligence , graph , matrix decomposition , blind signal separation , speech recognition , baseline (sea) , natural language processing , pattern recognition (psychology) , theoretical computer science , computer network , oceanography , geology , philosophy , linguistics , eigenvalues and eigenvectors , physics , channel (broadcasting) , quantum mechanics
Source separation is popular problem in which English datasets is used by default. Besides, source separation or speech enhancement is an important pre-processing step for following processes e.g. automatic speech recognition, automatic answering machine or hearing ads…However, experiments of source separation on Vietnamese dataset is quite modest as well as lack of Vietnamese standard datasets for source separation. To deal these issues, we build a Vietnamese dataset for source separation by collecting utterances of broadcasters from VTV’s official website. Moreover, a novel method was proposed by using sparse non-negative matrix factorization and graph regularization. Experiments showed that the proposed method is outperformed baseline.