Voice is one of the most promising tools in digital medicine. In association with virtual medical companions, the estimation of symptoms based on voice markers will allow both home monitoring of patients suffering from chronic neuropsychiatric diseases and access to personalized lifestyle advice for the general population. Sleepiness, present in many pathologies and presenting a very high prevalence both in patients suffering from chronic diseases and in the general population, is a privileged symptom for this approach. The objective of the work presented in this manuscript is thus to complete the information collected by virtual assistants during the interaction of the subjects with them, by using vocal markers validated as being reliable markers of sleepiness. The approach followed is the following. First, we introduce the mechanisms of voice production and the set of pathologies that can interfere with the different muscular and neuro-muscular functions involved, with a particular focus on the methodologies used for the recording and annotation of the corpora used. Next, we attempt to establish a consensus definition of sleepiness using three reference dictionaries of the French language; two text mining approaches~;~and finally through a general review of tools designed to measure it. We then present our own corpus of patients with hypersomnia, recorded at the university sleep medicine center of the Bordeaux University Hospital on a read-aloud task, annotated with both subjective (questionnaires) and objective (sleep latency on the Iterative Sleep Latency Test) measures of sleepiness validated by the physicians of the University Hospital. This corpus is then compared with other state-of-the-art corpora on the detection of sleepiness in voice, from which we propose recommendations on the development of such corpora. Then, using a perceptual study, we validate the use of the TILE database for the detection of sleepiness in speech. On the basis of this corpus, we develop four categories of vocal descriptors, measuring two dimensions of the impact of sleepiness on the voice. On the one hand, we study markers of acoustic voice quality; on the other hand, we design markers of reading quality, divided into three subcategories: reading errors made by patients, their automation through errors made by automatic speech recognition systems, and finally the durations and locations of reading pauses. These markers are validated on different forms of sleepiness (objective and subjective). Finally, we propose a methodology to train a classifier for the clinical use of these speech descriptors for the detection of three symptoms related to sleepiness. We propose a detailed analysis of the results obtained and the descriptors used by the classifier. To go further, we then propose to bring the classification problem closer to the reality of clinical reasoning by classifying two syndromes derived from the previous symptoms. Finally, in this same direction, we propose research perspectives around the symptom networks, in the framework of digital medicine research on sleepiness and on digital psychiatry in a more general way.

Virtual, here.

ID de réunion : 847 549 7370
Code secret : 9876