diff --git a/doc/source/release.rst b/doc/source/release.rst
index 926e8f1d0c5ea..e137a93af07a9 100644
--- a/doc/source/release.rst
+++ b/doc/source/release.rst
@@ -531,6 +531,7 @@ Bug Fixes
       names weren't strings (:issue:`4956`)
     - A zero length series written in Fixed format not deserializing properly.
       (:issue:`4708`)
+    - Fixed decoding perf issue on pyt3 (:issue:`5441`)
   - Fixed bug in tslib.tz_convert(vals, tz1, tz2): it could raise IndexError
     exception while trying to access trans[pos + 1] (:issue:`4496`)
   - The ``by`` argument now works correctly with the ``layout`` argument
diff --git a/pandas/io/pytables.py b/pandas/io/pytables.py
index 975d04c185d51..bc2f41502614d 100644
--- a/pandas/io/pytables.py
+++ b/pandas/io/pytables.py
@@ -3972,8 +3972,11 @@ def _unconvert_string_array(data, nan_rep=None, encoding=None):
     # where the passed encoding is actually None)
     encoding = _ensure_encoding(encoding)
     if encoding is not None and len(data):
-        f = np.vectorize(lambda x: x.decode(encoding), otypes=[np.object])
-        data = f(data)
+        try:
+            data = data.astype(str).astype(object)
+        except:
+            f = np.vectorize(lambda x: x.decode(encoding), otypes=[np.object])
+            data = f(data)
 
     if nan_rep is None:
         nan_rep = 'nan'