Mathieu Avanzi

Learn More
The paper has two goals: to present a toolbox for prosodic and phonostylistic description, and to use it for studying a specific radio style. This tool is quasi-automatic and modular. It consists of a set of Praat-based scripts like phonetic segmentation, melodic stylisation and prominence detection. It produces a phonostylistic report – called ProsoReport(More)
The aim of this paper is to present a tool developed in order to generate French rhythmical structure semi-automatically, without taking grammatical cues into account. On the basis of a phonemic alignment, the software first locates prominent syllables by considering basic acoustic features such as F0, duration and silent pause. It then assigns a degree of(More)
The automatic detection of prosodically prominent syllables is crucial for analysing speech, especially in French where prominence contributes substantially to prosodic grouping and boundary demarcation. In this paper, we compare different machine learning techniques for the automatic detection of prominent syllables, using prosodic features (including(More)
Up to now, prominence detection has mainly been considered a binary matter, a syllable or a word being considered as prosodically prominent or not. This contribution aims at developing an automatic detection procedure of gradual prominence. Based on 4 prosodic parameters (relative duration, relative f0, f0 movement and pause duration), the system provides(More)
This paper compares the prosody of 6 varieties of French spoken in three different areas: France (Paris and Lyon), Belgium (Tournai and Liège), and Switzerland (Geneva and Neuchâtel). The objective is to adress whether some regional varieties, namely those of Geneva and Tournai, are closer to standard French (i.e. the varieties spoken in France, represented(More)
The goal of this paper is to shed new light on accentuation in French, more precisely to discuss the role of grammatical constraints and of phonetic factors implicated in the perception of French final and non-final accent. The study is based on the analysis of a 70-minute long corpus, including various speaking styles. The corpus has been annotated(More)
In the area of large speech corpora, there is a definite need for common prosodic notation system based on efficient (semi)automating tools of prosodic segmentation and labelling. In this context, we present the software program ANALOR, developed in order to process semi-automatically prosodic data. From a text-sound alignment, this computer tool detects(More)
The aim of this paper is to present a software tool called ANALOR, which allows semi-automatic prominence detection in spontaneous French. On the basis of a manual annotation performed by two experts on a 70-minute long corpus including different regional varieties of French (Belgian, Swiss and metropolitan French) and various discourse genres (from read(More)
We present DisMo, a multi-level annotator for spoken language corpora that integrates part-of-speech tagging with basic disfluency detection and annotation, and multi-word unit recognition. DisMo is a hybrid system that uses a combination of lexical resources, rules, and statistical models based on Conditional Random Fields (CRF). In this paper, we present(More)