Proceedings of the KSPS conference (대한음성학회:학술대회논문집)

The Korean Society Of Phonetic Sciences And Speech Technology (대한음성학회)
권호 표지

Classification of pathological and normal voice based on dimension reduction of feature vectors

피처벡터 축소방법에 기반한 장애음성 분류

  • Lee, Ji-Yeoun (Speech and Audio Information Lab., Information and Communication Univ.) ;
  • Jeong, Sang-Bae (Speech and Audio Information Lab., Information and Communication Univ.) ;
  • Choi, Hong-Shik (Institute of Logopedics and Phoniatrics, Department of Otorhinolaryngology, Yongdong Severance Hospital, Yonsei University College of Medicine) ;
  • Hahn, Min-Soo (Speech and Audio Information Lab., Information and Communication Univ.)
  • 이지연 (한국정보통신대학교 음성음향정보연구실) ;
  • 정상배 (한국정보통신대학교 음성음향정보연구실) ;
  • 최홍식 (연세대학교 의과대학 영동세브란스병원 이비인후과학교실 음성언어의학연구소) ;
  • 한민수 (한국정보통신대학교 음성음향정보연구실)
  • Published : 2007.05.18
Download PDF
⟨ Previous Next ⟩

Abstract

This paper suggests a method to improve the performance of the pathological/normal voice classification. The effectiveness of the mel frequency-based filter bank energies using the fisher discriminant ratio (FDR) is analyzed. And mel frequency cepstrum coefficients (MFCCs) and the feature vectors through the linear discriminant analysis (LDA) transformation of the filter bank energies (FBE) are implemented. This paper shows that the FBE LDA-based GMM is more distinct method for the pathological/normal voice classification than the MFCC-based GMM.

Keywords