Alzheimer disease has considerable impact in speech. Recently several research works have proposed to use machine learning for early detection of the disease, mostly using discourse related features. In this master thesis the possibility of using additional acoustic features extracted from speech will be explored. Through a collaboration with a small company, recordings of speech from patients in different stages of the disease will be available for experimentation.
Goals: Find the acoustic parameters best suited for automatic Alzheimer detection.
Requirements: Basic knowledge of machine learning and some experience in python programming is mandatory.
Advisor: Eva Navas Cordón, Inma Hernáez Rioja