残響劣化した音声に対するノンレファレンス音声了解度推定における推定精度向上

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: 2023/08/01

タイトル(英語): On Improvement to Non-Reference Speech Intelligibility Estimation Accuracy for Reverberant Speech

著者名: 中澤　和司（山形大学大学院理工学研究科），近藤　和弘

著者名(英語): Kazushi Nakazawa (Yamagata University Graduate School of Science and Technology), Kazuhiro Kondo

キーワード: 音声了解度，深層学習，残響，音声強調，ノンレファレンス推定　　speech intelligibility，deep neural networks，reverberation，speech enhancement，non-reference estimation

要約(英語): In this paper, we made improvements and evaluated our proposed model for non-reference speech intelligibility estimation on reverberant speech, attempting to improve the estimation accuracy significantly. Our proposed method consists of a DNN for speech enhancement and a separate DNN for intelligibility estimation. The latter uses features obtained from enhanced and degraded speech to estimate intelligibility. Although previous studies have effectively estimated intelligibility for speech degraded by additive noise using similar models, they did not consider the degradation of distortion caused by reverberation. They also did not quantify the effect of various speech enhancement DNN models, the structure of the intelligibility prediction DNN, and the selection of parameters during feature calculation on estimation accuracy. Accordingly, we compared two top-of-the-line speech enhancement DNN models and used their output to train intelligibility prediction DNNs for reverberant speech while also varying the parameters used in the feature calculation. Consequently, the linear correlation coefficient between subjective and estimated intelligibility came to 0.801 with the best combination.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.143 No.8 （2023） 特集Ⅰ：Smart City を支える高度な情報通信，センシング及び医療技術　特集Ⅱ：情報処理技術の適用による有用情報の獲得と応用

本誌掲載ページ: 830-841 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/143/8/143_830/_article/-char/ja/