Any appreciable listener echo on telephone lines is perceived as undesirable, and considerable effort is expended to hold echo to acceptable levels. To provide a basis for appropriately controlling listener echo, a series of four subjective tests was conducted at Bell Laboratories to obtain subjective evaluations of the effects of listener echo on telephone transmission quality. The subjective tests included conditions in which the listener echo-path loss was flat or frequency-shaped by selective filtering. The test results showed that subjective opinions for conditions with the same singing margin (minimum value of echo-path loss) were highly dependent on the frequency shaping. A weighted echo-path loss (WEPL) is defined to provide a weighting on the frequency-shaped test conditions so that subjectively equivalent test conditions have approximately the same levels of WEPL. The results of these subjective tests are used to formulate a model of subjective opinion for use in network planning studies. The resulting listener echo opinion model is incorporated into an existing transmission rating model which encompasses loss, noise, and talker echo. Objectives are proposed for listener echo based on WEPL.
[1]
J. L. Sullivan.
A laboratory system for measuring loudness loss of telephone connections
,
1971
.
[2]
R. W. Hatch,et al.
Models for the subjective effects of loss, noise, and talker echo on telephone connections
,
1976,
The Bell System Technical Journal.
[3]
W. R. Daumer,et al.
A subjective comparison of selected digital codecs for speech
,
1978,
The Bell System Technical Journal.
[4]
F. P. Duffy,et al.
Speech signal power in the switched message network
,
1978,
The Bell System Technical Journal.
[5]
A. H. Inglis,et al.
An improved telephone set
,
1951,
Electrical Engineering.