音声の重なりに頑健な指定話者音声区間検出

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2015/08/01

タイトル(英語): Robust Extraction of Desired Speaker's Utterance in Overlapped Speech

著者名: 陸　昊澤（千葉大学大学院 融合科学研究科），赤岩　祐真（千葉大学大学院 融合科学研究科），堀内　靖雄（千葉大学大学院 融合科学研究科），黒岩　眞吾（千葉大学大学院 融合科学研究科）

著者名(英語): Haoze Lu (Graduate School of Advanced Integration Science, Chiba University), Yuma Akaiwa (Graduate School of Advanced Integration Science, Chiba University), Yasuo Horiuchi (Graduate School of Advanced Integration Science, Chiba University), Shingo Kuroiwa (Graduate School of Advanced Integration Science, Chiba University)

キーワード: 話者照合，混合音声モデル，指定話者音声区間検出　　speaker verification，overlapped speech model，extraction of desired speaker’s utterance

要約(英語): In this paper, we propose a speaker indexing method using speaker verification technique to extract one desired speaker's utterances from conversational speech. To solve the overlapped speech problem, we construct overlapped speech models with the observed conversational speech itself. The overlapped speech models include overlapped speech of target and cohort speaker, and speech model of two cohort speakers. In order to evaluate the proposed method, we made a simulated conversational speech that has up to 50% overlapping segments. The EER was reduced by up to 43.7% compared with the conventional methods that use a target speaker model only, and use a target model and overlapped speech model trained with a speaker independent large speech database.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.135 No.8 （2015） 特集：知能メカトロニクス分野と連携する知覚情報技術

本誌掲載ページ: 1009-1016 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/135/8/135_1009/_article/-char/ja/