In wordnet, a "synset" represents a meaning, and is in principle language-independent. All synsets have English-based names, but are linked to lemmas in multiple languages. You can get a list of all *lemmas* for a particular language, e.g. Farsi,
with wn.words(lang="fas")
Depending on what you are after, you could collect all synsets for which there is at least one Farsi lemma like this:
all_farsi_synsets = set(s for lemma in wn.words(lang="fas") for s in wn.synsets(lemma, lang="fas"))
As a side-note, it seems that there are some problems (i.e., bugs) with the mapping: Some of the synsets you will find in this way do not link back to a Farsi lemma.
Best,
Alexis