ATMOSPHERIC EFFECTS ON VOICE COMMAND INTELLIGIBILITY FROM ACOUSTIC HAIL AND WARNING DEVICES
Open Access
- Author:
- Bostron, Jason Henry
- Graduate Program:
- Acoustics
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- April 24, 2008
- Committee Members:
- Timothy A Brungart, Thesis Advisor/Co-Advisor
Timothy Edward Mc Devitt, Thesis Advisor/Co-Advisor - Keywords:
- intelligibility
outdoor sound propagation
voice command
variation
acoustic hail and warning device
AHWD - Abstract:
- Measurements of voice command sound pressure levels (SPLs) from acoustic hail and warning devices (AHWDs) were performed at listener locations up to 1500 m in order to determine how changes in meteorological condition affect variations in received voice command SPL. Variations in measured voice command SPL were compared to variations in meteorological condition, the results of other experiments, and the international standard for the attenuation of outdoor sound, ISO 9613-2. Similar to other experiments, our measurements indicate that variations of more than 30 dB should be expected to occur throughout the course of each day at listener locations at and beyond 200 m. However, unlike other experiments variations in measured voice command SPL of as much as 17 dB in a 6 minute time period were shown. The American standard for the calculation of the speech intelligibility index, ANSI S3.5, was used to calculate the intelligibility of measured voice commands in order to determine how intelligibility at a listener location varies as a function of meteorological condition. When a moderate background noise level was present, a change in measured voice command SPL of only 3–4 dB was necessary to cause a partially intelligible voice command to become either fully intelligible or unintelligible. This result is significant because measured voice command SPLs were shown to vary by 30 dB at a listener location of 200 m due to changes in meteorological condition. The intelligiblity rating at a listener location can be estimated using a constant signal-to-noise ratio (S/N) for signal SPLs below 60 dB. However, for speech signals above 60 dB, the S/N must be increased to ensure the same intelligibility rating. This result is significant because AHWDs are commonly broadcast into high background noise level environments and/or into high insertion loss enclosures where signal SPLs above 60 dB must be used.