Presentation 2016-01-28
Validation of the Effects of Ensemble Learning for i-vector-based Speaker Identification
Shohei Sonoda, Masato Inoue,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Currently, most speaker identification methods have been performed by i-vectors which represent the features of unique speakers. Since i-vectors of utterances are supposed to be distributed on a surface of a hypersphere, it is possible to identify the speaker by using cosine similarity of i-vectors. In this manuscript, we suggest a new approach employing weak learners that significantly contribute to the field of machine leaning to improve the accuracy of the recognition in recent years. Especially, ensemble learning methods such as Bagging and Random Forest are examined and validated.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Speaker Identification / i-vector / Ensemble Learning / Random Forest / Bagging
Paper # NC2015-58
Date of Issue 2016-01-21 (NC)

Conference Information
Committee NC / NLP
Conference Date 2016/1/28(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kyushu Institute of Technology
Topics (in Japanese) (See Japanese page)
Topics (in English) Implementation of Neuro Computing,Analysis and Modeling of Human Science, etc
Chair Toshimichi Saito(Hosei Univ.) / Kenya Jinno(Nippon Inst. of Tech.)
Vice Chair Shigeo Sato(Tohoku Univ.) / Naoto Fujisaka(Hiroshima City Univ.)
Secretary Shigeo Sato(Kyushu Inst. of Tech.) / Naoto Fujisaka(Kyoto Sangyo Univ.)
Assistant Hiroyuki Kanbara(Tokyo Inst. of Tech.) / Hisanao Akima(Tohoku Univ.) / Hidehiro Nakano(Tokyo City Univ.) / Hiroyuki Asahara(Okayama Univ. of Science)

Paper Information
Registration To Technical Committee on Neurocomputing / Technical Committee on Nonlinear Problems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Validation of the Effects of Ensemble Learning for i-vector-based Speaker Identification
Sub Title (in English) Bagging vs Random forest
Keyword(1) Speaker Identification
Keyword(2) i-vector
Keyword(3) Ensemble Learning
Keyword(4) Random Forest
Keyword(5) Bagging
1st Author's Name Shohei Sonoda
1st Author's Affiliation Waseda University(Waseda Univ)
2nd Author's Name Masato Inoue
2nd Author's Affiliation Waseda University(Waseda Univ)
Date 2016-01-28
Paper # NC2015-58
Volume (vol) vol.115
Number (no) NC-426
Page pp.pp.13-16(NC),
#Pages 4
Date of Issue 2016-01-21 (NC)