diff --git a/spacy/tests/regression/test_issue768.py b/spacy/tests/regression/test_issue768.py
new file mode 100644
index 000000000..d8c8be80b
--- /dev/null
+++ b/spacy/tests/regression/test_issue768.py
@@ -0,0 +1,36 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from ...language import Language
+from ...attrs import LANG
+from ...fr.language_data import TOKENIZER_EXCEPTIONS, STOP_WORDS
+from ...language_data.punctuation import TOKENIZER_INFIXES, ALPHA
+
+import pytest
+
+
+@pytest.fixture
+def fr_tokenizer_w_infix():
+    SPLIT_INFIX = r'(?<=[{a}]\')(?=[{a}])'.format(a=ALPHA)
+
+    # create new Language subclass to add to default infixes
+    class French(Language):
+        lang = 'fr'
+
+        class Defaults(Language.Defaults):
+            lex_attr_getters = dict(Language.Defaults.lex_attr_getters)
+            lex_attr_getters[LANG] = lambda text: 'fr'
+            tokenizer_exceptions = TOKENIZER_EXCEPTIONS
+            stop_words = STOP_WORDS
+            infixes = TOKENIZER_INFIXES + [SPLIT_INFIX]
+
+    return French.Defaults.create_tokenizer()
+
+
+@pytest.mark.parametrize('text,expected_tokens', [("l'avion", ["l'", "avion"]),
+                                                  ("j'ai", ["j'", "ai"])])
+def test_issue768(fr_tokenizer_w_infix, text, expected_tokens):
+    """Allow zero-width 'infix' token during the tokenization process."""
+    tokens = fr_tokenizer_w_infix(text)
+    assert len(tokens) == 2
+    assert [t.text for t in tokens] == expected_tokens