Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

Voice Personality Transformation Using an Orthogonal Vector Space Conversion

Ki Seung Lee, Dae Hee Youn, Il Whan Cha

Department of Electronics Engineering, Yonsei University, Sudaemoon-Ku, Seoul, Korea

A newly developed voice personality transformation algorithm is introduced in this paper. Voice personality transformation is the process of changing one person's acoustic features (source) to those of another person (target). In this paper, personality transformation is achieved by changing the LPC cepstrum coefficients, excitation spectrum and global/local pitch contour. An orthogonal vector space conversion technique is proposed to transform LPC the cepstrum coefficients. This technique consists of principle component decomposition by applying the Karhunen-Loeve(KL) transformation and minimum mean-square error coordinate transformation(MSECT). To transform prosodic characteristics, we propose a simple pitch contour modification method. The experimental results show the effectiveness of the proposed algorithm in both subjective and objective evaluations.

Full Paper

Bibliographic reference.  Lee, Ki Seung / Youn, Dae Hee / Cha, Il Whan (1995): "Voice personality transformation using an orthogonal vector space conversion", In EUROSPEECH-1995, 427-430.