Could you please advise on the process

The names should be avaiable as metadata if you used the default ingestion pipeline or you can add them as as metadata when creating the nodes.

I am storing the chroma db in a file once embeddings are generated. From stored chromadb collection I want to know the documents available in it

vector_store = self.setup_chroma(collectionName)
index = VectorStoreIndex.from_vector_store(
vector_store,
service_context=self.service_context,
)

I think youd just have to use chromadbs API directly -- I think they have a get_all() function, but would have to read their docs