diff --git a/spacy/training/converters/conll_ner_to_docs.py b/spacy/training/converters/conll_ner_to_docs.py
index 3b851039c..902db585b 100644
--- a/spacy/training/converters/conll_ner_to_docs.py
+++ b/spacy/training/converters/conll_ner_to_docs.py
@@ -2,9 +2,9 @@ from wasabi import Printer
 
 from .. import tags_to_entities
 from ...training import iob_to_biluo
-from ...lang.xx import MultiLanguage
 from ...tokens import Doc, Span
 from ...util import load_model
+from ...util import load_model, get_lang_class
 
 
 def conll_ner_to_docs(
@@ -86,7 +86,7 @@ def conll_ner_to_docs(
     if model:
         nlp = load_model(model)
     else:
-        nlp = MultiLanguage()
+        nlp = get_lang_class("xx")()
     output_docs = []
     for conll_doc in input_data.strip().split(doc_delimiter):
         conll_doc = conll_doc.strip()
@@ -136,7 +136,7 @@ def segment_sents_and_docs(doc, n_sents, doc_delimiter, model=None, msg=None):
             "Segmenting sentences with sentencizer. (Use `-b model` for "
             "improved parser-based sentence segmentation.)"
         )
-        nlp = MultiLanguage()
+        nlp = get_lang_class("xx")()
         sentencizer = nlp.create_pipe("sentencizer")
     lines = doc.strip().split("\n")
     words = [line.strip().split()[0] for line in lines]