The Use of Transformed Normal Speech Data in Acoustic Model Training for Non-Audible Murmur Recognition

• Item type: SIG Technical Reports(1)
• 公開日: 2011-01-28
• タイトル: The Use of Transformed Normal Speech Data in Acoustic Model Training for Non-Audible Murmur Recognition
• タイトル / 言語: en
• タイトル: The Use of Transformed Normal Speech Data in Acoustic Model Training for Non-Audible Murmur Recognition
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: 音響モデル
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Nara Institute of Science and Technology
• 著者所属: Nara Institute of Science and Technology
• 著者所属: Nara Institute of Science and Technology
• 著者所属: Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者名: Denis, Babani; Tomoki, Toda; Hiroshi, Saruwatari; Kiyohiro, Shikano
• 著者名(英): Denis, Babani; Tomoki, Toda; Hiroshi, Saruwatari; Kiyohiro, Shikano
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: This paper presents a novel approach to the acoustic model training for Non-Audible Murmur (NAM) recognition using normal speech data transformed into NAM data. NAM is extremely soft murmur, which is so quiet that people around the speaker hardly hear it. NAM recognition is one of the promising silent speech interfaces for man-machine speech communication. Our previous work has shown the effectiveness of Speaker Adaptive Training (SAT) based on Constrained Maximum Likelihood Linear Regression (CMLLR) in the NAM acoustic model training. However, since the amount of available NAM data is still small, the effect of SAT is limited. In this paper we propose modified SAT methods capable of using a larger amount of normal speech data by transforming them into NAM data. The transformation of normal speech data is performed with the CMLLR adaptation. The experimental results demonstrate that the proposed methods yield an absolute increase of around 2% in word accuracy compared with the conventional method.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: This paper presents a novel approach to the acoustic model training for Non-Audible Murmur (NAM) recognition using normal speech data transformed into NAM data. NAM is extremely soft murmur, which is so quiet that people around the speaker hardly hear it. NAM recognition is one of the promising silent speech interfaces for man-machine speech communication. Our previous work has shown the effectiveness of Speaker Adaptive Training (SAT) based on Constrained Maximum Likelihood Linear Regression (CMLLR) in the NAM acoustic model training. However, since the amount of available NAM data is still small, the effect of SAT is limited. In this paper we propose modified SAT methods capable of using a larger amount of normal speech data by transforming them into NAM data. The transformation of normal speech data is performed with the CMLLR adaptation. The experimental results demonstrate that the proposed methods yield an absolute increase of around 2% in word accuracy compared with the conventional method.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10442647
• 書誌情報: 研究報告 音声言語情報処理（SLP）; 巻 2011-SLP-85, 号 2, p. 1-6, 発行日 2011-01-28
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会