add from_language

app/document_processor.py

@@ -103,7 +103,7 @@ def process_documents(docs):
         encode_kwargs={"normalize_embeddings": True}
     )
     
-    text_splitter = RecursiveCharacterTextSplitter(
+    text_splitter = RecursiveCharacterTextSplitter.from_language("id")(
         chunk_size=1500,
         chunk_overlap=300
     )