Dear all,

I wonder whether anybody knows a Turkish corpus / lexical database which shows the stress of lexical items. I am aware of TELL project; however, stress is not encoded in the current version.

Currently I am working on stress pattern of lexical items in Turkish; unfortunately, the corpora I know do not indicate any information about stress. I am trying to compile data by using a dictionary; however, I realized that information about stress there is not complete and also not consistent. I would be really happy to hear any suggestions / ideas.

Thanks in advance!

Similar questions and discussions