DocumentCode
3451914
Title
Glottal Pulse Shape Optimization using Simulated Annealing
Author
Bahaadini, Sara ; Sameti, Hossein ; Jabbari, Fattaneh ; Mohammadi, Seyed Hamidreza
Author_Institution
Dept. of Comput. Eng., Sharif Univ. of Technol., Tehran, Iran
fYear
2012
fDate
2-3 May 2012
Firstpage
112
Lastpage
115
Abstract
Excitation signal has essential role in speech synthesis filters to produce natural speech. In this study, a new method is proposed for modeling the glottal pulse shape of a speaker. A search is done on the glottal pulse shape space using simulated annealing method. The PESQ measure and Cepstral distance between the original signal and the synthesized signal are used as the cost function. An LPC filter with 10 coefficients is used as the synthesis filter. The PESQ value between the original and synthesized speech using traditional impulse is 2.402. Here, the glottal pulse for a certain speaker is modeled in three different experiments. In the first experiment, the negative PESQ measure is used as the cost function. The PESQ value of 2.881 is achieved in this experiment. In the second experiment, the Cepstral distance is used as cost function, resulting in the PESQ value of 2.697. In the third experiment, only the vowel parts are considered in computing the distance measure, which results in the PESQ value of 2.733.
Keywords
filtering theory; simulated annealing; speech synthesis; Cepstral distance; LPC filter; PESQ measure; excitation signal; glottal pulse shape optimization; natural speech; simulated annealing; speaker glottal pulse shape; speech synthesis filters; Cepstral analysis; Cost function; Hidden Markov models; Shape; Simulated annealing; Speech; Speech synthesis; PESQ; Simulated Annealing; pulse shape; speech quality; speech synthesis filter;
fLanguage
English
Publisher
ieee
Conference_Titel
Artificial Intelligence and Signal Processing (AISP), 2012 16th CSI International Symposium on
Conference_Location
Shiraz, Fars
Print_ISBN
978-1-4673-1478-7
Type
conf
DOI
10.1109/AISP.2012.6313728
Filename
6313728
Link To Document