Reconstruction of Normal Speech from Whispered Speech based on BP Neural Network
Lists
Only you see this
Abstract
This paper tells of a new approach for reconstructing normal speech from Chinese whispered speech based on BP neural network.First,the formants of normal speech and whispered speech are acquired,and the BP neural network is used to get a model of the conversion from the whispered speech to the formants of normal speech,then the formant of whispered speech is converted by adopting this model.Finally,the whispered speech is converted into the normal speech by using the formant synthesizer.Simulation results show that the score of the DRT of the converted speech is 80% and the MOS is 3.5,both intelligibility and quality of the converted speech are satisfied.
BibTeX
@inproceedings{han2009reconstruction,
author = {Tao Han},
title = {{Reconstruction of Normal Speech from Whispered Speech based on BP Neural Network}},
booktitle = {Communications},
year = {2009},
}Loading related content...