Influence of the operations of the middle and inner ear was examined after recording transiently evoked otoacoustic emissions (TEOAEs) to a click stimulus in guinea pigs by using the ILO88. After opening of auditory bulla, TEOAE power under 2kHz in FFT pictures (FEP-2) was increased significantly. On the other hand, TEOAE power between 2 and 4kHz in FFT pictures (FEP2-4) did not change after the operation. The filling of physiological saline in the bulla developed disappearance of TEOAE powers reversibly. Total echo power was decreased slowly after destruction of the cochlear apical turn. To be precise, FEP-2 was decreased more quickly, and then FEP2-4 decreased. TEOAE power above 4kHz was decreased finally. These results implied that click evoked otoacoustic emission is frequency specific to some extent.
For the evaluation of transmission efficiency of speech discrimination test, an articulation score is generally used. The articulation score, however, contains no specific information on what syllable are pronounced about speech recognition in patients with sensorineural hearing loss. Therefore, we analyzed that data obtained from speech discrimination test by transient matrices. Furthermore, we analyzed speech discrimination ability of patients with sensorineural hearing loss by application of index of information loss which is called “conditional entropy”. The conditional entropy is the probability of joint occurrence of input and output for information. Patients with sensorineural hearing loss of age groups of 10-34 years, 35-64 years and over 65 years were analyzed separately. Analysis of information loss was efficient to examine the recognition performance for consonant and vowel parts. The pattern of confusion in transient matrices was more prominent in the over 65 years of age group than the younger patient groupes. Furthermore, in analysis of information loss at +40dB discrimination, the over 65 years group showed as twice equivocation as much that of the 10-34 years group.
The benefit of compression on hearing aids is to decrease distortion by the limit of maximum output. In this paper we investigated the relation between the compression ratio of a full dynamic range compression (FDRC) hearing aid “ED2”, speech sound level, and speech perception (words distinction). “ED2” was applied to 20 ears of 11 persons. The compression ratio was set on 1:1, 1.5:1 and 3:1. The stimulus level of speech sounds varied from 50dBA, 65dBA to 80dBA. Each test of speech sound perception consisted of 25 different words. The following results were obtained. When the level of speech sounds was 50dBA, the average of words distinction ratio was 43.1% in the condition of compression ratio of 1:1, 77.6% in 1.5:1, and 84.3% in 3:1 respectively. At 65dBA, they were 78.8% in 1:1, 90.8% in 1.5:1, and 93.6% in 3:1 respectively. At 80dBA, they were 89.2% in 1:1, 93.0% in 1.5:1, and 90.1% in 3:1. These differences in the word distinction were statistically significant (T test) in compression ratio between 1:1 and 1.5:1 at 50dBA (p=.0001), and between 1:1 and 1.5:1 at 65dBA (p=.0155). These results suggested the effectiveness of FDRC on hearing aids.
In this study, a time-variant adaptive digital filter was designed for the purpose of improving the articulation scores of Japanese consonant-vowell syllables in the hearing impaired. In the system, some band-pass filters were generated beforehand which match the human auditory filters. The local peaks of consonant spectrum in high frequency were selectively detected by the running spectrum method, and the proper filter characteristics were adopted to the monosyllables to enhance transient formants and band noise of consonants. Articulation scores and phonetic confusion among the original and modified syllables were experimentally investigated with the pseudo hearing impaired. As a result, modified plosives improved the articulation, however, modified fricatives increased phonetic confusion. Other monosyllables decreased confusion, but exhibited no significant improvement. These results suggest that enhancing the local peaks of consonant is effective to the consonants where formant transients are used as a psychoacoustic cue in auditory perception.
Two kinds of simplified word discrimination tests have been developed to assist in determining whether a young hearing-impaired child can perceive spectral fine aspects or only prosodic gross acoustic patterns of speech for his/her auditory comprehension. The Japanese auditory number test required the child with optimum hearing-aid to identify counted sequences and individual numbers using audio-tape-cards. It was administered to 76 children of ages of 4-13 years old with mean hearing-levels (HL) at 70-133dB. As the results, discrimination scores were related to mean HL, and generally the scores were high for mean HL better than 90dB and low for HL poor than 113dB. Although a range from minimum to maximum scores were obtained for children with audiogram at approximately 90-110dB. The Japanese family terms discrimination test employed common word lists: okaasan (mother), otoosan (father), oneesan (sister), oniisan (brother), obaasan (grandmother), ojiisan (grandfather), obasan (aunt), ojisan (uncle), papa (papa), mama (mama). The test were performed in 381 severe to profound hearing-impaired children of 5 years old enrolled in 66 schools for the deaf all over Japan. It was found that they achieved more than 80% in their discrimination scores for “okaasan, ” “otoosan, ” and “oneesan” in the auditory-visual (speech-reading) conditions. For the other word lists, they achieved more than 70% in their discrimination scores. In the visual-only conditions, the average scores were 18% to 29% down from the previous scores. By constructing the lists with 10 easy words whose spectral and/or prosodic information is critical for their discrimination, several useful perspectives of the evaluatdon of auditory skills of hearingimpaired children were obtained. Particularly, for the word group “oniisan”, “ojiisan” and “ojisan, ” when confusion matrices within the word groups were investigated, a distinctive cue in their auditory perception would be defined by the simple test materials. These tests are recommended as clinical evaluation of the child's perceptual abilities in a short time to guide his/her continuing aural (re) habilitation program.
To determine the optimal stimulus condition to induce distortion-product otoacoustic emissions (DPOAEs), DPOAEs (2f1-f2) were recorded under various stimulus levels of f1 (4.0kHz) and f2 (4.8kHz) ranging from 20 to 70dBSPL in 10 guinea pigs. When the sound pressure level of f1 and/or f2 was equal to or less than 60dBSPL, DPOAEs were completely abolished by anoxia lasting within 5 minutes. DPOAEs induced by the greater sound pressure, however, were reduced slowly and reached to the certain levels in one hour which was almost eqivalent to the levels recorded within the calibration box. It is most likely that the anoxia sensitive response reflects the active process of the cochlea and non-sensitive one is induced by the passive process of the cochlea and/or mechanical distortion of the recording system. The largest amplitude of anoxia sesitive DPOAEs was obtained at each sound pressure level of f1 when f1 was larger than f2 by 10dB. The conclusion is that the maximum response of DPOAEs, related directly to the active process and uncontaminated with the response induced by other phenomena, is obtained when f1 and f2 are 60dBSPL and 50dBSPL, respectively.