Paper Info
Title
Building HMM based unit-selection speech synthesis system using synthetic speech naturalness evaluation score
Abstract
This paper proposes a unit-selection and waveform concatenation speech synthesis system based on synthetic speech naturalness evaluation. A Support Vector Machine (SVM) and Log Likelihood Ratio (LLR) based synthetic speech naturalness evaluation system was introduced in our previous work. In this paper, the evaluation system is improved in three aspects. Finally, a unit-selection and concatenation waveform speech synthesis system is built on the base of the synthetic speech naturalness evaluation system. Optimum unit sequence is chosen through the re-scoring for the N-best path. Subjective listening tests show the proposed synthetic speech evaluation based speech synthesis system significantly outperforms the traditional unit-selection speech synthesis system.
Year
DOI
Venue
2011
10.1109/ICASSP.2011.5947567
ICASSP
Keywords
Field
DocType
kernel fisher discriminant,llr,unit-selection concatenation speech synthesis system,svm,waveform concatenation speech synthesis system,speech synthesis,support vector machine,unit-selection speech synthesis,synthetic speech naturalness evaluation score,hidden markov models,log likelihood ratio,hmm based unit-selection speech synthesis system,synthetic speech evaluation,context modeling,speech,support vector machines,acoustics,hidden markov model,context model
Speech synthesis,Likelihood-ratio test,Pattern recognition,Computer science,Naturalness,Support vector machine,PSQM,Context model,Speech recognition,Artificial intelligence,Concatenation,Hidden Markov model
Conference
Volume
Issue
ISSN
null
null
1520-6149 E-ISBN : 978-1-4577-0537-3
ISBN
Citations 
PageRank 
978-1-4577-0537-3
3
0.67
References 
Authors
5
4
Name
Order
Citations
PageRank
Heng Lu1589.22
Zhen-Hua Ling285083.08
Li-Rong Dai31070117.92
Ren-Hua Wang434441.36