I am not familiar with the nuances surrounding fuzzy string matching for the Chinese language. Having said that, you can use pre-trained embeddings to perform the string matching:
from polyfuzz import PolyFuzz
from polyfuzz.models import Embeddings
from flair.embeddings import WordEmbeddings
chinese_embedding = WordEmbeddings('zh')
matcher = Embeddings(chinese_embedding, min_similarity=0)
model = PolyFuzz(matcher)
I am not familiar with the nuances surrounding fuzzy string matching for the Chinese language. Having said that, you can use pre-trained embeddings to perform the string matching: