Title :
Speech Synthesis Research Based on EGG
Author :
Lijiang Chen ; Xia Mao ; Pengfei Wei ; Compare, Angelo
Author_Institution :
Sch. of Electron. & Inf. Eng., Beihang Univ., Beijing, China
Abstract :
Objective: In this paper, Electroglottograph (EGG) is adopted to improve the naturalness of low bit-rate formant speech synthesis. Methods: EGG inverting waveform is used as glottal excitation of the formant speech synthesis. A new SUV divided method combining with EGG and speech signal is recommend, to increase the accuracy of the extraction of formant. Results: According to the subjective MOS score and comparative experiment with the original speech signal, the naturalness of voice formant synthesis combined EGG is considerable higher than the traditional use of the pulse signal and the inverse filtered signal as the glottal excitation. Conclusion: The combination of EGG formant speech synthesis improves the naturalness of synthetic speech.
Keywords :
feature extraction; speech synthesis; waveform analysis; EGG formant speech synthesis; EGG inverting waveform; MOS score; SUV divided method; electroglottograph; formant extraction; glottal excitation; low bit-rate formant speech synthesis; speech signal; synthetic speech naturalness; voice formant synthesis naturalness; Educational institutions; Filtering; Speech; Speech synthesis; Synthesizers; Vibrations; EGG; Formant; Speech Synthesis;
Conference_Titel :
Green Computing and Communications (GreenCom), 2013 IEEE and Internet of Things (iThings/CPSCom), IEEE International Conference on and IEEE Cyber, Physical and Social Computing
Conference_Location :
Beijing
DOI :
10.1109/GreenCom-iThings-CPSCom.2013.332
