Frequency Warping for Speaker Adaption of Text-to-speech Synthesis
Vocal tract length normalization (VTLN) is generally used in speech recognition for removing individual speaker characteristics. In this paper, we employ VTLN to speaker adaptation of speech synthesis. We propose a new frequency warping approach to reduce the spectrum distance between source and target speakers. The frequency warping function is based on a bilinear function and the warping factor is dynamically generated frame-by-frame. The warped spectra of source speaker are then converted to LSPs to train hidden Markov models (HMM). HMMs are further adapted by maximum likelihood linear regression (MLLR) with target speakers data. The experimental results show that our frequency warping approach can make the warped spectra of source speaker closer to target speaker and the resultant adapted HMMs have a better performance than the HMMs trained with unwarped spectra in term of voice naturalness and speaker similarity.
frequency warping speaker adaptation TTS
Weixun Gao Qiying Cao
School of Information Science and Technology, Donghua Univeristy, Shanghai, China Shanghai Normal Un College of Computer Science & Technology, Donghua Univeristy, Shanghai, China
国际会议
北京
英文
307-310
2010-09-26（万方平台首次上网日期，不代表论文的发表时间）