Effects of masking noise on vowel and sibilant contrasts in normal-hearing speakers and postlingually deafened cochlear implant users
Schematic illustration of hypotheses for normal-hearing speakers (panel A) and cochlear implant users (panel B) recorded at and post-implant.
AVS (A-mels), Duration (D-ms) and SPL (S-dB) vs. noise-to-signal ratio (dB) for each of the seven speakers who had normal hearing. For each panel, values of AVS are on the left vertical axis; the range of durations is shown by numbers in the upper and lower right-hand corners; values of SPL are on the right vertical axis. Values of N/S were obtained by multiplying extracted S/N by and binning the resulting values into seven equally-spaced intervals. The values on the x-axis are located at the centers of the class intervals. Results from the quiet conditions are shown separately at the left of each plot (N/S is arbitrary.) Subject designations: ; . hearing. Error bars: standard error about the mean.
Data averaged across six speakers with normal hearing (values in standard deviations). To compensate for inter-subject differences in data averages and ranges, each subject’s data were standardized and the standardized values were averaged across subjects. See caption for Fig. 2 for remaining details. Data for Subject FNH7 were not included.
AVS (A-mels), Duration (D-ms) and SPL (S-dB) vs. noise-to-signal ratio (N/S-dB) for each of the 7 implant users. There are two plots for each implant user, an upper one of data from one month and a lower one from one year post-implant. Subject designations: . For further detail, see caption of Fig. 2 .
Data averaged across the seven implant users, at one-month (panel A) and one-year (panel B) post-implant. See caption of Fig. 3 for remaining details.
Sibilant contrast distance (C), spectral mean for /s/ (S) and spectral mean for /ʃ/ (H) vs N/S (standardized values, as for Figs. 3 and 5 ). Results for speakers with normal hearing are shown in panel A, for implant users at one-month in panel B and for one-year post-implant in panel C. (For further details, see captions of Figs. 3 and 5 .)
Schematic diagram illustrating a possible explanation of the results. Average vowel spacing vs. normalized N/S. The solid function (NH) shows the result for the speakers with normal hearing; the dashed function (Cl-Yr), for the implant users at one year post-implant, and the dotted line (CI-Mo), for the implant users at one month.
Summary of ANOVA results. of variance accounted for by the effect. (eta-squared) is calculated from ( Young, 1993 ). NL: Noise level.
Article metrics loading...
Full text loading...