Objective Detection of the Speech Frequency Following Response (sFFR): A Comparison of Two Methods

Area of Science:

• Auditory Neuroscience
• Speech Processing
• Signal Detection Theory

Background:

• Speech frequency following responses (sFFRs) are crucial for understanding auditory processing.
• Automated detection methods for sFFRs can enhance efficiency in clinical and research settings.
• Development of robust stopping rules for sFFR recordings is needed.

Purpose of the Study:

• To compare the detection accuracy and time efficiency of two statistical methods for automated sFFR detection: the F-test and Hotelling's T² (HT²) test.
• To evaluate the robustness of these automated detection methods in quiet and speech-shaped noise conditions.
• To assess the potential interchangeability of the F-test and HT² test for sFFR analysis.

Main Methods:

• sFFRs were recorded from 18 normal-hearing adult listeners.
• Data were collected in both quiet and speech-shaped noise conditions.
• The F-test and HT² test were applied to automatically detect sFFRs, and their performance was analyzed.

Main Results:

• Both the F-test and HT² test demonstrated similar detection accuracy for sFFRs across quiet and noise conditions.
• The HT² test exhibited less variability in test time compared to the F-test.
• Automated sFFR detection proved robust under the tested conditions.

Conclusions:

• Statistically-based automated sFFR detection using either the F-test or HT² test is reliable for auditory responses in quiet and speech-shaped noise.
• The HT² test offers more consistent timing, potentially beneficial for clinical applications.
• Further research is recommended to validate these methods with diverse auditory stimuli and masking conditions.