The influence of prosody on segmental phenomena such as French schwa and liaison has been pointed out, but never systematically studied on speech data. In this paper we propose a comprehensive coding system to be used on a large oral corpus. Such a system aims at providing explicit and robust data which can serve as a solid base to test the different claims which have been put forward in the literature.

Text available on https://www.isca-speech.org/archive/archive_papers/interspeech_2005/i05_1437.pdf .