I am trying to come up with a way to divide French words into syllables, phonemes, morphemes, or any other individually pronounceable/meaningful sub-unit in order to model word corruption in expressive aphasia.


I am looking for a way to automate the process of dividing words into syllables. This can be an algorithm that implements some rule system of syllabification, or a pre-made lookup table of french words along with their syllables in some clear format (something like élément -> é.lé.ment; méchant -> mé.chant; ...).

What I could find:

I was able to download syllable counts from this Wiktionary page, which was enormously helpful, but it would be much better if I can use the syllabic form of the word directly. My research could not yield a publicly available dataset for French syllables (or phonemes, morphemes), nor the code for any algorithm that can be used to generate them (or a public API for one of the online tools that do).


2 Answers 2


After 2 more days of research, I stumbled upon PyHyphen, a python package that does exactly what I want (not only for french). I also checked lexique.org after @Lambie's comment. It does contain the information I was looking for but PyHyphen's format is more suitable for my use case.


Since your concern is ultimately with human speech and not text processing, you might want to look at speech-centric research, which focuses on what facts of language indicate a particular parsing of phonemes into syllables. You could start with the work of Roland Noske (the link is to his publication list which contains many links to downloadable papers), as well as the works of François Dell (e.g. his book Generative Phonology and French Phonology). There is also a volume of Lingua edited by Durand & Hintze on French phonology. The reason for looking at the linguistic literature on French phonology is that one aspect of expressive aphasia is the ability to speak (of course the relevance of linguistic syllabification ultimately depends on what exact data you are trying to model). These references will also direct you to other aspects of French phonology, such as the phonemes and their featural analysis, plus the linguistic discussion can serve as a corrective on text-processing tools such as hyphenation dictionaries.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.