def test_parse_cmu(self): test_str = b"""\ ;;; some test data to ensure that parsing CMU-formatted files works ADOLESCENT AE2 D AH0 L EH1 S AH0 N T ADOLESCENT(1) AE2 D OW0 L EH1 S AH0 N T """ cmufh = io.BytesIO(test_str) pronunciations = pronouncing.parse_cmu(cmufh) self.assertTrue(len(pronunciations) > 0) matches = [x for x in pronunciations if x[0] == 'adolescent'] self.assertEqual(len(matches), 2)