Many articles have used voice analysis to detect Parkinson’s disease (PD), but few have focused on the early stages of the disease, the impact of telephonic recordings, and the gender effect. We have adapted the latest speaker recognition system, called x-vectors, in order to detect PD at an early stage using voice analysis. X-vectors are embeddings extracted from Deep Neural Networks (DNNs), which provide robust speaker representations and improve speaker recognition when large amounts of training data are used.
We recorded 221 French speakers (recently diagnosed PD subjects and healthy controls) with a high-quality microphone and via the telephone network. We assessed the impact of the audio segment durations, data augmentation, type of dataset used for the neural network training, kind of speech tasks, and back-end analyses. X-vectors technique provided better classification performances than classical Gaussian Mixture Models for the text-independent tasks, and seemed to be particularly suited for the early detection of PD in women (7 to 15% improvement).
Curriculum vitæ
Dijana Petrovska-Delacrétaz obtained her degree in Physics and her PhD from the Swiss Federal Institute of Technology (EPFL) in Lausanne. She was working as a consultant at AT&T, as a post-Doc at Télécom ParisTech, and as a Senior Scientist in the Informatics Department of Fribourg University, Switzerland. Since 2004 she joined Télécom SudParis as Associate Professor. Her research activities are oriented towards pattern recognition, signal processing, and data-driven machine learning methods, that are exploited for different applications such as speech, speaker and language recognition, very low-bit speech coding, biometrics (2D and 3D face, and voice), crypto-biometrics (including privacy preserving biometrics), and health applications.
