Kindle2Txt is a simple program that will convert Kindle Word lookup data into a simple text file.
The run parameters look like this:
Kindle2Txt.py ([vocab_db_path]) (lang) (output_folder)
An example first run for only japanese books with the output saved in the current directory would be:
Kindle2Txt.py "d:\system\vocabulary\vocab.db" ja .
As long as the vocab.db path never changes, you won't need any parameters after the first run. It will keep track of the last sentence timestamp that was output, so additional runs should only output new lookups since the last run.