diff --git a/doc/whats_new/v0.22.rst b/doc/whats_new/v0.22.rst
index 1221571e1f658..3d96489c48a72 100644
--- a/doc/whats_new/v0.22.rst
+++ b/doc/whats_new/v0.22.rst
@@ -308,6 +308,10 @@ Changelog
   removes accents from strings that are in NFKD normalized form. :pr:`15100` by
   :user:`Daniel Grady <DGrady>`.
 
+- |Fix| Fixed a bug that caused :class:`feature_extraction.DictVectorizer` to raise
+  an `OverflowError` during the `transform` operation when producing a `scipy.sparse`
+  matrix on large input data. :pr:`15463` by :user:`Norvan Sahiner <norvan>`.
+
 :mod:`sklearn.feature_selection`
 ................................
 
diff --git a/sklearn/feature_extraction/_dict_vectorizer.py b/sklearn/feature_extraction/_dict_vectorizer.py
index 6e820f93a3b60..ca49263f57913 100644
--- a/sklearn/feature_extraction/_dict_vectorizer.py
+++ b/sklearn/feature_extraction/_dict_vectorizer.py
@@ -154,7 +154,7 @@ def _transform(self, X, fitting):
         X = [X] if isinstance(X, Mapping) else X
 
         indices = array("i")
-        indptr = array("i", [0])
+        indptr = [0]
         # XXX we could change values to an array.array as well, but it
         # would require (heuristic) conversion of dtype to typecode...
         values = []
@@ -182,7 +182,6 @@ def _transform(self, X, fitting):
             raise ValueError("Sample sequence X is empty.")
 
         indices = np.frombuffer(indices, dtype=np.intc)
-        indptr = np.frombuffer(indptr, dtype=np.intc)
         shape = (len(indptr) - 1, len(vocab))
 
         result_matrix = sp.csr_matrix((values, indices, indptr),