Human voice as a measure of mental load level

Abstract

Purpose: The aim of this study was to determine a reliable with speech to attest variations in mental load level during and efficient set of acoustic parameters of the human voice the experimental task. able to estimate individuals’ mental load level. Implementing Results: Classic features (fundamental frequency, its detection methods and real-time analysis of mental load is standard deviation, number of periods) and original a major challenge for monitoring and enhancing human features (frequency modulation and short-term variation task performance, especially during high-risk activities (e.g., in digital amplitude length) of the acoustic signals were flying aircraft). predictive of memory load condition. They varied significantly Method: The voices of 32 participants were recorded during according to the number of words to recall, specifically a cognitive task featuring word list recall. The difficulty of beyond a threshold of 3–5 words to recall, that is, when the task was manipulated by varying the number of words memory performance started to decline. in each list (i.e., between 1 and 7, corresponding to 7 mental Conclusions: Some acoustic parameters of the human load conditions). Evoked pupillary response, known to be a voice could be an appropriate and efficient means for useful proxy of mental load, was recorded simultaneously detecting mental load levels.