Date of Award
Fall 1979
Document Type
Thesis - Restricted
Degree Name
Master of Science (MS)
Department
Electrical and Computer Engineering
First Advisor
Niederjohn, Russell J.
Second Advisor
Jeutter, Dean C.
Third Advisor
Moeller, Arthur C.
Abstract
Enhancement of the intelligibility of speech in noise is a complex and unsolved problem. In order to gain some understanding of the speech in noise situation, the relative importance of various frequency ranges to the intelligibility of speech in noise is investigated in this thesis. Some conclusions which are drawn for the frequency range and S/N ratios used include: l) No high- pass or low-pass filter increases the intelligibility of speech in noise. 2) Speech can be low-pass or high-pass filtered at 2500Hz without significant loss of intelligibility. 3) The intelligibility of speech decreases rapidly with increasing high-pass filter cutoff frequency. 4) With speech in noise, listeners tend to set the level of the composite signal based upon the level of the speech, regardless of the S/N ratio. The Articulation Index is calculated and validated for the experimental data gathered in this work.
Recommended Citation
Mliner, David G., "The Measurement and Prediction of the Effect of Low-Pass and High-Pass Filtering Upon the Intelligibility of Speech in White Noise" (1979). Master's Theses (1922-2009) Access restricted to Marquette Campus. 4727.
https://epublications.marquette.edu/theses/4727