Results of speech processor upgrade in a population of Veterans Affairs cochlear implant recipients.

OBJECTIVE: This study aimed to evaluate the change in performance when the Speak processing strategy replaced the Mpeak strategy. SUBJECTS: Twenty-one veterans who participated in the original Department of Veterans Affairs Multicenter Cochlear Implant Study, received the Nucleus F0F1F2 processor (Cochlear Corp., Englewood, CO, U.S.A.) (WSP) and were subsequently upgraded to the Mpeak (MSP) processing strategy. METHODS: Closed- and open-set speech recognition in quiet and in noise in auditory only conditions at the phoneme, word, and sentence levels were evaluated. RESULTS: Changes in audiologic parameters were evaluated using a paired t-test for related samples. Statistically significant improvements in mean scores were found for all tests and were greatest for the speech-in-noise conditions. An examination of individual data revealed that 61% of the subjects improved for tests administered in quiet, whereas up to 69% of the subjects had improved performance in noise. Subjects with no open set speech discrimination using the Mpeak strategy showed no improvement with the Speak strategy. CONCLUSIONS: Cochlear implants can provide improved performance using modifications to external speech processor hardware and software without changing the internal implant.

Low- and high-pass filtered Northwestern University Auditory Test No. 6 for monaural and binaural evaluation.

Two protocols developed for the Tonal and Speech Materials for Auditory Perceptual Assessment audio compact disc are low- and high-pass filtered versions of the Northwestern University Auditory Test No. 6 (female speaker). Two experiments were performed on subjects with normal hearing to determine the needed filter characteristics of the words. In Experiment I, word-recognition performance was determined at 70 dB SPL for four low-pass cut-offs (800 to 1700 Hz) and for four high-pass cut-offs (1700 to 3000 Hz). Based on the data from Experiment I, the low-pass (1500-Hz cut-off) and high-pass (2100-Hz cut-off) conditions were selected for inclusion on the compact disc. In Experiment II, psychometric functions were obtained in quiet for the two filter conditions with lists 3 and 4 included on the compact disc. Because the high-pass filtered words were recorded on channel 1 and concurrently the corresponding low-pass filtered words were recorded on channel 2, a binaural fusion paradigm is also available on the compact disc. For the compact disc trials, which involved 120 subjects with normal hearing, psychometric functions were generated for monaural low-pass, monaural high-pass, and binaural spectral fusion (low- and high-pass) conditions, with each of the subjects providing data at one of six levels for each of three conditions.

Multiple frequency tympanometry: effects of ear canal volume compensation on static acoustic admittance and estimates of middle ear resonance.

Three methods for compensating multiple frequency acoustic admittance measurements for ear canal volume were studied in 26 men with normal middle ear transmission systems. Peak compensated static acoustic admittance (magnitude of y) and phase angle (phi) were calculated from sweep frequency tympanograms (226-1243 Hz in 113 Hz increments). Of the procedures used to compensate for volume in rectangular form, the ear canal pressure used to estimate volume had the largest effect on the estimate of middle ear resonance. Median resonance was 800 Hz for admittance measurements compensated at 200 daPa versus 1100 Hz for measurements compensated at -350 daPa. The remaining two methods, compensation of susceptance only versus both susceptance and conductance and compensation using the minimum volume versus separate volumes at each frequency, did not affect estimates of middle ear resonance. Estimates of middle ear resonance from compensated phase angle measurements also were compared with estimates of resonance from admittance and phase difference curves. Although resonance could not be estimated from the phase difference curve, resonance estimated from the admittance difference curve agreed with the estimate from compensated phase angle.

Spondaic word detection and recognition functions for female and male speakers.

The Department of Veterans Affairs recently produced a compact disc of speech audiometry materials. The compact disc, which is available commercially, includes the W-1 spondaic words recorded by a female speaker. Two experiments were conducted. The purposes of experiment 1 were to obtain normative detection and recognition data on the female recording of the spondaic words and to compare the detection and recognition functions for the original male speaker version of the W-1 words. No significant differences were found between the recognition functions for each speaker. The recognition functions for both speakers were displaced to higher sound-pressure levels by 8 dB above the detection functions. Clinically, the two versions of the W-1 spondaic words should produce equivalent results. In experiment 2, slopes of the individual spondaic word recognition functions for the female speaker were obtained from two listeners and are discussed in terms of interstimulus, intertrial, and intersubject variability.