Install required libs:
conda install ffmpeg pydub nltk spacy
pip install -r requirements.txt
https://github.com/sebastiandziadzio/ivr-synsets Last working version of Słowosieć: 2.3.7
First download wordnet corpus using interactive window (run from Python shell):
import nltk
nltk.download()
Then download Słowosieć (PlWordNet) from:
http://www.nlp.pwr.wroc.pl/plwordnet/download/plwordnet_2_3.7z
Overwrite the files from *_pwn_format dir (path could be checked in the NLTK download window) (tip: make backup of original files)
Download corpus for spacy:
python -m spacy download pl_core_news_sm
Example could be run using:
python nlp.py