In addition, if one refers to each of the PEAQ in-
ternal model output variables (MOVs), the statistical
correlation in some cases is also near the 85% mark,
as indicated in Table 4.
Table 4: Statistical correlation ρ between MOS and PEAQ
(basic and advanced) internal MOVs.
PEAQ Basic ρ PEAQ Adv. ρ
MOV MOV
1 0.02 1 −0.82
2 0.02 2 −0.73
3 −0.51 3 −0.50
4 −0.79 4 0.46
5 −0.61 5 −0.65
6 0.46 - -
7 −0.75 - -
8 −0.84 - -
9 −0.71 - -
10 −0.36 - -
11 −0.55 - -
The MOS attributed to each test sentence, or-
ganized by increasing order, is shown in Figure 8.
Furthermore, it includes the corresponding objec-
tive grades yielded by the PESQ method, since they
strongly correlate with MOS, as indicated in Table 3.
10 20 30 40 50 60 70 80
1
2
3
4
5
Test signal index (sorted by MOS)
Rating
MOS
PESQ
Figure 8: MOS and PESQ grades for all test signals.
Overall, the results shown in Tables 3 and 4 as
well as in Figure 8 suggest that the both the PESQ
and the PEAQ standards may provide a starting point
to the development of an effective objective method
for quality assessment of audioband speech signals
degraded by reverberation.
5 CONCLUSIONS
This paper addressed the problem of quality evalu-
ation of audioband (24 kHz) speech signals with re-
spect to the reverberation effect. Mathematical mod-
els were reviewed and the most important reverber-
ation aspects for the application at hand were indi-
cated. Subjective listening tests were designed and
performed to quantify via MOS the human percep-
tion of speech impairment by reverberation. Corre-
lation between objective and subjective quality mea-
sures have been computed in order to verify the po-
tential ability of standard quality-evaluation methods
in predicting the subjective quality of speech signals
spoiled by reverberation.
REFERENCES
Allen, J. B. (1982). Effects of small room reverberation on
subjective preference. J. Acoustic. Soc. Am., 71:S5.
Allen, J. B. and Berkley, D. A. (1979). Image method for
efficiently simulating small-room acoustics. J. Acous-
tic. Soc. Am., 65(4):943–950.
de Lima, A. A., Freeland, F. P., de Jesus, R. A., Bispo, B. C.,
Biscainho, L. W. P., Netto, S. L., Said, A., Kalker, A.,
Schafer, R., Lee, B., and Jam, M. (2008). On the qual-
ity assessment of sound signals. In Proc. IEEE Int.
Symp. on Circuits and Systems. pp. 426–429, Seattle,
USA.
Gardner, W. G. (1992). The virtual acoustic room. Mas-
ter’s thesis, School of Architecture and Planning, MIT,
Cambridge, USA.
Gardner, W. G. (1998). Applications of Digital Signal Pro-
cessing, chapter Reverberation Algorithms, pp. 85–
131. Kluwer, New York, USA.
ITU-R Rec. BS.1387 (1998). Method for objective mea-
surements of perceived audio quality. ITU.
ITU-T Rec. G.191 (2005). Software tools for speech and
audio coding standardization. ITU.
ITU-T Rec. P.563 (2004). Single-ended method for ob-
jective speech quality assessment in narrow-band tele-
phony applications. ITU.
ITU-T Rec. P.862 (2001). Perceptual evaluation of speech
quality (PESQ): Objective method for end-to-end
speech quality assessment of narrow band telephone
networks and speech codecs. ITU.
ITU-T Rec. P.862.2 (2005). Wideband extension to rec-
ommendation P.862 for the assessment of wideband
telephone networks and speech codecs. ITU.
Jot, J.-M. and Chaigne, A. (1991). Digital delay networks
for designing artificial reverberators. Presented at the
90th Convention of the AES. Preprint 3030.
Perkins, M. E., Dvorak, C. A., Lerich, B. H., and Zebarth,
J. A. (1999). Speech transmission performance plan-
ning in hybrid IP/SCN networks. IEEE Commun.
Mag., 37(7):126–131.
Schroeder, M. R. (1962). Natural sounding artificial rever-
beration. J. Audio Eng. Soc., 10(3):219–233.
Schroeder, M. R. (1965). New method of measuring rever-
beration time. J. Acoustic. Soc. Am., 37:409–412.
Tan, C., Moore, B. C. J., Zacharov, N., and Mattila, V.
(2004). Predicting the perceived quality of nonlinearly
distorted music and speech signals. J. Audio Eng. Soc.,
52(7/8).
SIGMAP 2008 - International Conference on Signal Processing and Multimedia Applications
262