Statistical Parametric Speech Synthesis Based on Speaker and Language Factorization
Authors:
- 7 February 2012
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech, and Language Processing
- Vol. 20 (6) , 1713-1724
Abstract
An increasingly common scenario in building speech synthesis and recognition systems is training on inhomogeneous data. This paper proposes a new framework for estimating hidden Markov models on data containing both multiple speakers and multiple languages. The proposed framework, speaker and language factorization, attempts to factorize speaker-/language-specific characteristics in the data and then model them using separate transforms. Language-specific factors in the data are represented by transforms based on cluster mean interpolation with cluster-dependent decision trees. Acoustic variations caused by speaker characteristics are handled by transforms based on constrained maximum-likelihood linear regression. Experimental results on statistical parametric speech synthesis show that the proposed framework enables data from multiple speakers in different languages to be used to: train a synthesis system; synthesize speech in a language using speaker characteristics estimated in a different language; and adapt to a new language.Keywords
This publication has 27 references indexed in Scilit:
- Context adaptive training with factorized decision trees for HMM-based statistical parametric speech synthesisSpeech Communication, 2011
- Statistical parametric speech synthesisSpeech Communication, 2009
- A Speech Parameter Generation Algorithm Considering Global Variance for HMM-Based Speech SynthesisIEICE Transactions on Information and Systems, 2007
- New approach to the polyglot speech generation by means of an HMM-based speaker adaptable synthesizerSpeech Communication, 2006
- Solving unsymmetric sparse systems of linear equations with PARDISOFuture Generation Computer Systems, 2004
- A compact model for speaker-adaptive training2002
- Speech parameter generation algorithms for HMM-based speech synthesis2002
- Rapid speaker adaptation in eigenvoice spaceIEEE Transactions on Speech and Audio Processing, 2000
- Cluster adaptive training of hidden Markov modelsIEEE Transactions on Speech and Audio Processing, 2000
- Maximum likelihood linear transformations for HMM-based speech recognitionComputer Speech & Language, 1998