HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

This paper introduces novel paradigms for the segmentation of speech into syllables. The main idea of the proposed method is based on the use of a time-frequency representation of the speech signal, and the fusion of intensity and voicing measures through various frequency regions for the automatic selection of pertinent information for the segmentation. The time-frequency representation is used to exploit the speech characteristics depending on the frequency region. In this representation, intensity profiles are measured to provide information into various frequency regions, and voicing profiles are measured to determine the frequency regions that are pertinent for the segmentation. The proposed method outperforms conventional methods for the detection of syllable landmark and boundaries on the TIMIT database of American-English, and provides a promising paradigm for the segmentation of speech into syllables.

This paper presents SLAM : a simple method for the automatic Stylization and LAbelling of speech Melody. The main contributions over existing methods are : the alphabet of melodic contours is fully data-driven, an explicit time-frequency representation is used to derive complex melodic contours, and melodic contours can be determined over arbitrary prosodic/syntactic units. Additionally, the system can handle some specificities of spontaneous speech (e.g., multi speakers, speech turns and speech overlaps). A preliminary experiment conducted on 3 hours of spoken French indicates that a small number of contours is sufficient to explain most of the observed contours. The method can be easily adapted to other stressed languages. The implementation is open-source and freely available † .