Daniel Pressnitzer

This paper detects sleep deprivation using voice analysis with machine learning, identifying two distinct vocal biomarkers: prosody and timbre changes, which vary individually.

The paper presents a model of auditory frequency change perception, showing that listeners combine spectral fine structure and envelope cues adaptively based on sound type and individual differences.

The paper introduces a method to interpret machine-learning classifiers by analyzing how noise and reverse correlation affect classification decisions across various data types and models.