Frequency-importance and transfer functions for recorded CID W-22 word lists

Area of Science:

• Audiology
• Speech Science
• Acoustics

Background:

• The Articulation Index (AI) is a tool used to predict speech intelligibility.
• Existing AI standards may not fully capture the characteristics of specific speech tests like the CID W-22.
• Accurate AI calculation requires precise frequency-importance and transfer functions.

Purpose of the Study:

• To derive and report frequency-importance and transfer functions for the Technisonic Studios' CID W-22 word test recordings.
• To enable the calculation of AI values and prediction of W-22 test scores using these new functions.
• To compare the derived functions with existing standards.

Main Methods:

• Utilized word recognition scores from 8 normal-hearing listeners.
• Tested listeners under 308 conditions involving filtering and masking.
• Derived frequency-importance and transfer functions from the collected data.

Main Results:

• Reported novel frequency-importance and transfer functions specifically for the CID W-22 test.
• The derived importance function exhibits a broader frequency range and different shape compared to the current ANSI standard (1969).
• The transfer function shows a similar slope to the ANSI transfer function for 256 PB-words but is shifted by 0.05 AI.

Conclusions:

• The developed functions provide a more accurate method for calculating AI and predicting W-22 test performance.
• These findings suggest a need to update or supplement existing AI standards for specific speech materials.
• The study contributes to improved audiological assessments and speech intelligibility predictions.