DocumentCode :
1998262
Title :
Feature analysis for quality assessment of reverberated speech
Author :
De Lima, Amaro A. ; De Prego, T.M. ; Netto, Sergio L. ; Lee, Bowon ; Said, Amir ; Schafer, Ronald W. ; Kalker, Ton ; Fozunbal, Majid
Author_Institution :
PEE/COPPE, Fed. Univ. of Rio de Janeiro, Rio de Janeiro, Brazil
fYear :
2009
fDate :
5-7 Oct. 2009
Firstpage :
1
Lastpage :
5
Abstract :
This paper analyzes the ability of several measurements to quantify the reverberation effect in speech signals. We consider an intrusive scheme, in which the clean and reverberated signals are available, allowing one to estimate the corresponding room impulse response (RIR) signal. An artificial neural network (ANN) is trained for all features and used in a regression approach to estimate the human perceptual evaluation in a mean opinion score (MOS) 1-5 scale. Dimensionality reduction approaches are applied to generate a simpler ANN regression, establishing the most representative features for the problem at hand. A correlation level of 85% with subjective test scores was achieved by reducing the input-vector dimension from 10 to 3, including only the features of reverberation time, room spectral variance, and direct-to-reverberant energy ratio.
Keywords :
neural nets; regression analysis; reverberation; speech processing; ANN regression; artificial neural network; feature analysis; human perceptual evaluation; input-vector dimension; intrusive scheme; mean opinion score; quality assessment; reverberated speech; reverberation effect; room impulse response signal; speech signals; Acoustic reflection; Artificial neural networks; Convolution; Humans; Performance analysis; Quality assessment; Reverberation; Signal analysis; Speech analysis; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia Signal Processing, 2009. MMSP '09. IEEE International Workshop on
Conference_Location :
Rio De Janeiro
Print_ISBN :
978-1-4244-4463-2
Electronic_ISBN :
978-1-4244-4464-9
Type :
conf
DOI :
10.1109/MMSP.2009.5293326
Filename :
5293326
Link To Document :
بازگشت