Accuracy of predicted ear canal speech levels using the VIOLA input/output-based fitting strategy

Area of Science:

• Audiology
• Hearing Aid Technology
• Signal Processing

Background:

• The Visual Input/Output Locator Algorithm (VIOLA) is a software-assisted method for prescribing hearing aid amplification targets.
• It assumes that hearing aids matching coupler targets will produce predicted speech levels in the ear canal.
• The validity of this assumption requires empirical investigation.

Purpose of the Study:

• To evaluate the accuracy of predicted ear canal speech levels generated by the VIOLA procedure.
• To determine if hearing aids selected using VIOLA meet prescribed amplification targets in a realistic ear canal environment.
• To identify factors contributing to discrepancies between predicted and actual amplified speech levels.

Main Methods:

• Six hearing aids (linear, compression, single/2-channel) were tested on a KEMAR manikin simulating hearing loss.
• Each hearing aid was programmed to VIOLA prescription targets.
• Continuous speech was presented, and amplified levels were measured in the manikin's ear canal with an unvented earmold.

Main Results:

• Predicted and observed ear canal speech levels showed good agreement for soft speech inputs.
• Discrepancies increased with higher speech input levels, with loud speech showing the largest differences.
• Observed levels were consistently lower than predicted levels for louder inputs.

Conclusions:

• The VIOLA procedure shows promise for controlling nonlinear hearing aid parameters.
• Deviations between predicted and observed levels were attributed to hearing aid limiting, compression, and individual acoustic characteristics.
• Modifications to VIOLA software to account for limiting and compression ratio effects could improve accuracy, particularly for loud speech inputs.