Ideally, this project should involve somebody from India who knows about Gujarati language.
There is a sample pdf file having text in Gujarati attached with this project that you should have a look at. You need to convert it into xml file, and filter it out for ten words that we will supply you after selecting for the project. You should share the info/ method of doing it with us in its entirety. This is a simple task that should not take much of your time, so please bid accordingly.
There is a chance that it might lead you into being invited for a much larger assignment.