Prosodic Parameters and Prosodic Structures of French Emotional Data

Katarina Bartkova (1) , Denis Jouvet (2) , Elisabeth Delais-Roussarie (3)

1 ATILF - Analyse et Traitement Informatique de la Langue Française
2 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
3 LLF UMR7110 - Laboratoire de Linguistique Formelle

The detection and modelling of emotions in speech remains a challenging issue in speech processing. The aim of the study presented here is to analyze and compare the use of several prosodic parameters in emotional speech in French. The data set used for the study contains utterances recorded in six emotional types: anger, fear, sadness, disgust, surprise and joy. The sentences of the emotional data are also recorded by the same speaker in a neutral reading style allowing a comparison between emotional and neutral speech. The prosodic analysis focuses to the main prosodic parameters such as vowel duration, energy and F0 level, and pause occurrences. The values of prosodic parameters are compared among the various emotional styles, as well as between emotional style and neutral style utterances. Moreover, the structuration of the sentences, in the various emotional styles, is particularly studied here through a detailed analysis of pause occurrences and their length, and of the length of prosodic groups.

Mots clés: emotional speech; emotions; neutral style; prosodic parameters

Domaines: Traitement du signal et de l'image [eess.SP]

Katarina Bartkova, Denis Jouvet, Elisabeth Delais-Roussarie. Prosodic Parameters and Prosodic Structures of French Emotional Data. Speech Prosody 2016, May 2016, Boston, United States. ⟨hal-01293516⟩