The Effects of Input-Output Configuration in Syllabic Compression on Speech Perception

Speech perception was tested through a broad-band syllabic compressor with four different static input-output configurations. All other parameters of the compressor were held constant. The compressor was implemented digitally and incorporated a delay to reduce overshoot. We studied four different input-output configurations, including a linear reference condition. Normal-hearing and hearing-impaired subjects participated in the experiments testing perception of meaningful sentences as well as nonsense CVCs in carrier phrases. The speech materials were presented in quiet and in noise. The results from the CVCs were analyzed quantitatively in terms of scores and qualitatively in terms of phoneme confusions. Differences in speech perception due to the different input-output configurations were small. The input-output configuration with the highest amplification of low amplitude sounds yielded the best results. Detailed analysis of the results included a correlational analysis with a number of auditory functions characterizing the ears tested. The pure-tone audiogram provided parameters of auditory sensitivity: average audiometric loss and audiometric slope. Psychophysical tests provided parameters of temporal resolution and frequency selectivity: the temporal resolution factor, temporal gap detection, and auditory filter shape. The correlational analysis showed that the subjects with better temporal acuity obtained better results.

KEY WORDS: compression, consonant confusions, temporal resolution, frequency selectivity, speech intelligibility

Submitted on December 6, 1990
Accepted on May 22, 1991

CiteULike    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?