How to manage sound, physiological and clinical data of 2500 dysphonic and dysarthric speakers?

Abstract

International audienceThe aim of this contribution is to propose a database model designed for the storage and accessibility of various speech disorder data including signals, clinical evaluations and patients' information. This model is the result of 15 years of experience in the management and the analysis of this type of data. We present two important French corpora of voice and speech disorders that we have been recording in hospitals in Marseilles (MTO corpus) and Aix-en-Provence (AHN corpus). The population consists of 2500 dysphonic, dysarthric and control subjects, a number of speakers which, as far as we know, constitutes currently one of the largest corpora of " pathological " speech. The originality of this data lies in the presence of physiological data (such as oral airflow or estimated sub-glottal pressure) associated with acoustic recordings. This activity led us to raise the question of how we can manage the sound, physiological and clinical data of such a large quantity of data. Consequently, we developed a database model that we present here. Recommendations and technical solutions based on MySQL, a relational database management system, are discussed

Assessment of impairment of intelligibility and of speech signal after oral cavity and oropharynx cancer

“…Autrement dit, il s'agit du degré avec lequel le message d'un locuteur peut être compris par un auditeur, soit la « proportion de parole comprise », ou, lors d'une évaluation, le « taux de mots correctement retranscrits » [3,4,5]. Elle dépend donc d'un jugement d'autrui, ce qui induit une difficulté de mesure, en lien avec les stratégies de compensation humaines permettant un meilleur décodage du signal grâce à la structure de la langue et le sens du message [6].…”

Perceptual Evaluation of Dysphonic Voices: Can a Training Protocol Lead to the Development of Perceptual Categories?

“…Our final choice was the reading aloud of Alphonse Daudet's ''La ch evre de Monsieur Seguin'' (Mr. Seguin's Goat), which has been used for about 20 years to record dysphonic patients at the Ear, Nose, and Throat Ward of the Timone University Hospital in Marseille. 22 Given that dysphonia can manifest itself temporarily and in a nonuniform way, we chose the sentence ''Il les perdait toutes de la même façon'' (He was losing all of them in the same way) because it contains a series of voiced/voiceless transitions and vocalic continuums, and its prosodic structure is such that the word ''toutes'' (all) is stressed. This structure is particularly interesting because the combination of an F0 rise (which is not always realized properly by dysphonics) and a (voiceless stop + vowel + voiceless stop + vowel) syllable structure is likely to reveal cases of laryngeal difficulty.…”

“…The database includes samples from about 2500 dysphonic or dysarthric speakers. 22 For our categorization task, we confined the corpus to voices of women with various types of dysfunctional dysphonia resulting from nodules, polyps, cysts, or Reinke's edema. Speakers with no vocal disorders were added, giving us a selection of 400 female voices.…”

Dysarthria in individuals with Parkinson's disease: a protocol for a binational, cross-sectional, case-controlled study in French and European Portuguese (FraLusoPark)

“…Owing to inter-speaker variability, any generalisation drawn from speech parameters in the clinical population requires data from a large number of speakers 87. The FraLusoPark project is in line with this idea in that it builds a large-scale corpus of PD speech recordings and includes a large set of metadata (clinical examinations, speech measurements, linguistic features, patient-based indices).…”