Development And Application Of Mandarin Materials For Speech Audiometry

[Objective] The purpose of this study was to psychometrically equate two sets of phonetically balanced Mandarin bisyllabic word lists, establish norms for disproportionate loss in speech intelligibility, summarize the effect of hearing rehabilitation in cochlear implant users. [Test one] 964 bisyllabic words were digitally recorded by male talker of standard Mandarin,352 words were selected to compose 10 lists. Percentage of correct word recognition was measured for each word at four intensity levels using 20 normally hearing subjects. The order of the presentation of the lists was randomized for each subject. Using Statistica7.0 Performance-Intensity function for each word was plotted, slopes and thresholds of them were calculated.242 words were chosen after that. The thresholds and slopes of these words were in Guassian distribution. [Test two] Digitally adjusted these words and included them in two sets of phonetically balanced Mandarin bisyllabic word lists. One set include six lists of 40 words each. The other set include four lists of 40 words each, the intensity is down with 5dB between every 5 words. For the first set of lists 12 subjects with normal hearing served in our equivalent test. Four presentation levels were selected for evaluation. The order of the presentation was randomized for each subject. Original data were expressed as percentile scores. Two-factor ANOVA was used to compare the recognition scores of the 6 lists. These lists were considered to be primarily equivalent with each other. For the other set of lists 36 subjects with normal hearing served in our equivalent test. The order of the presentation was randomized for each subject. Original intensity was given according to Spearman-Karber and results were expressed as thresholds. Two-factor ANOVA was used to compare the thresholds of the 4 lists. These lists were considered to be primarily equivalent with each other.[Test three] We included the results from both ears of the 165 patients with sensorineural hearing loss. All patients had been evaluated on puretone audiometry and speech audiometry. PBmax was abtained by monosyllable. We examined the correlation among the variables of puretone audiometry and concluded that PBmax correlates best with PTA3.We draw a scatterplot of all 165 PBmax scores as a function of PTA3. Furthermore if we draw a straight line along the lower age of the data points that approximately 98% of the points lie above the line. Any PBmax score falling below this boundary may be considered, with high probability to be disproportionately poor. [Test four] Free field measurement, speech audiometry were performed in 24 cochlear implant users. Speech audiometry include sentence intelligibility, monosyllable lists, bisyllabic word lists.The test works out in a listening and repeating way. Original data were expressed as percentile scores. Sound field audiometry showed that their hearing threshold reached 38.38 dB HL, the speech discrimination score for sentence was 80%, for monosyllable was from 0%-80%, the threshold for bisyllabic word lists was 59.14 dB HL. Most cochlear implant users have clearer pronunciation.