diff --git a/qurator/dinglehopper/character_error_rate.py b/qurator/dinglehopper/character_error_rate.py
index e99f391..29826e3 100644
--- a/qurator/dinglehopper/character_error_rate.py
+++ b/qurator/dinglehopper/character_error_rate.py
@@ -6,6 +6,7 @@ from typing import Tuple
 from uniseg.graphemecluster import grapheme_clusters
 
 from qurator.dinglehopper.edit_distance import distance
+from qurator.dinglehopper.ocr_files import ExtractedText
 
 
 def character_error_rate_n(reference, compared) -> Tuple[float, int]:
@@ -14,12 +15,13 @@ def character_error_rate_n(reference, compared) -> Tuple[float, int]:
 
     :return: character error rate and length of the reference
     """
+    if isinstance(reference, str):
+        return character_error_rate_n(
+                ExtractedText.from_text(reference),
+                compared)
+
     d = distance(reference, compared)
-    # XXX
-    from .cli import ExtractedText
-    if isinstance(reference, ExtractedText):
-        reference = reference.text
-    n = len(list(grapheme_clusters(unicodedata.normalize('NFC', reference))))
+    n = len(list(grapheme_clusters(reference.text)))
 
     if d == 0:
         return 0, n
diff --git a/qurator/dinglehopper/edit_distance.py b/qurator/dinglehopper/edit_distance.py
index 284b676..bc607a9 100644
--- a/qurator/dinglehopper/edit_distance.py
+++ b/qurator/dinglehopper/edit_distance.py
@@ -7,6 +7,7 @@ from typing import Sequence, Tuple
 import numpy as np
 from uniseg.graphemecluster import grapheme_clusters
 
+from .ocr_files import ExtractedText
 
 def levenshtein_matrix(seq1: Sequence, seq2: Sequence):
     """Compute the matrix commonly computed to produce the Levenshtein distance.
@@ -75,12 +76,12 @@ def distance(s1, s2):
     Note that this is different from levenshtein() as this function knows about Unicode normalization and grapheme
     clusters. This should be the correct way to compare two Unicode strings.
     """
-    # XXX
-    from .cli import ExtractedText
+
     if isinstance(s1, ExtractedText):
         s1 = s1.text
     if isinstance(s2, ExtractedText):
         s2 = s2.text
+
     s1 = list(grapheme_clusters(unicodedata.normalize('NFC', s1)))
     s2 = list(grapheme_clusters(unicodedata.normalize('NFC', s2)))
     return levenshtein(s1, s2)