Phonogenres, or speaking styles, are typified acoustic images associated to types of language activities, causing prosodic and phonostylistic variations. This communication presents a large speech corpus (7 hours) in French, extending a previous work by Goldman et al. (2011a), Simon et al. (2010), with a greater number and complementary repertoire of considered phonogenres. The corpus is available with segmentation at phonetic, syllabic and word levels, as well as manual annotation. Segmentations and annotations were achieved semi-automatically, through a set of Praat implemented tools, and manual steps. The phonogenres are also described with a reduced set of situational dimensions as in Lucci (1983) and Koch & Oesterreicher’s (2001). A preliminary acoustic study, joining rhythmical comparative measurements (Dellwo 2010) to Goldman et al.’s (2007a) ProsoReport, reports acoustic differences between phonogenres.