File size: 101,006 Bytes
6fa4bc9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 |
{
"paper_id": "2021",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T01:13:04.103129Z"
},
"title": "Towards a morphological transducer and orthography converter for Western Tlacolula Valley Zapotec",
"authors": [
{
"first": "Jonathan",
"middle": [
"N"
],
"last": "Washington",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "Swarthmore College",
"location": {
"addrLine": "500 College Ave. Swarthmore",
"postCode": "19081",
"region": "PA",
"country": "USA"
}
},
"email": "jonathan.washington@swarthmore.edu"
},
{
"first": "Felipe",
"middle": [
"H"
],
"last": "Lopez",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Danielle",
"middle": [],
"last": "Brook",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "Haverford College",
"location": {
"addrLine": "370 Lancaster Ave. Haverford",
"postCode": "19072",
"region": "PA"
}
},
"email": ""
},
{
"first": "",
"middle": [],
"last": "Lillehaugen",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "Haverford College",
"location": {
"addrLine": "370 Lancaster Ave. Haverford",
"postCode": "19072",
"region": "PA"
}
},
"email": ""
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "This paper presents work towards a morphological transducer and orthography converter for Dizhsa, or San Lucas Quiavin\u00ed Zapotec, an endangered Western Tlacolula Valley Zapotec language. The implementation of various aspects of the language's morphology is presented, as well as the transducer's ability to perform analysis in two orthographies and convert between them. Potential uses of the transducer for language maintenance and issues of licensing are also discussed. Evaluation of the transducer shows that it is fairly robust although incomplete, and evaluation of orthographic conversion shows that this method is strongly affected by the coverage of the transducer.",
"pdf_parse": {
"paper_id": "2021",
"_pdf_hash": "",
"abstract": [
{
"text": "This paper presents work towards a morphological transducer and orthography converter for Dizhsa, or San Lucas Quiavin\u00ed Zapotec, an endangered Western Tlacolula Valley Zapotec language. The implementation of various aspects of the language's morphology is presented, as well as the transducer's ability to perform analysis in two orthographies and convert between them. Potential uses of the transducer for language maintenance and issues of licensing are also discussed. Evaluation of the transducer shows that it is fairly robust although incomplete, and evaluation of orthographic conversion shows that this method is strongly affected by the coverage of the transducer.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "In this paper, we present work towards a morphological transducer and orthography converter for Dizhsa, also known in the academic literature as San Lucas Quiavin\u00ed Zapotec (SLQZ), an endangered language variety of Western Tlacolula Valley Zapotec [zab] . 1 To our knowledge, this is the first computational implementation of the morphology of a Zapotec language. (Throughout the paper we use the term \"language variety\" in place of \"dialect\" because of the pejorative force of the word dialecto in Spanish.)",
"cite_spans": [
{
"start": 247,
"end": 252,
"text": "[zab]",
"ref_id": null
},
{
"start": 255,
"end": 256,
"text": "1",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "A morphological transducer, implemented as a finite-state transducer (FST), is a tool that performs morphological analysis (converts between a word form and a morphological analysis) and morphological generation (the reverse). For example, a form like gunydir\u00ebng 'they won't do' can be quickly converted to an analysis like uny<v><tv><irre><neg>",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The three authors recognise that we live and work on the homeland of the Lenape, and pay respect and honor to the caretakers of this land, from time immemorial until now, and into the future.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "+\u00ebng<prn><pers><p3><prox><pl> (read as the negative irrealis form of the transitive verb whose stem is \"uny\", followed by a 3rd person proximal plural personal pronominal enclitic); similarly, the analysis can be quickly converted to the form.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Not all speakers of SLQZ write their language, though more and more are doing so (Lillehaugen, 2016) . There are published proposals for two orthographies, which we refer to as the phonemic orthography (Munro & Lopez et al., 1999) and the simple orthography (Munro et al., 2021) . An orthography converter between these two orthographies based on the morphological transducer has been developed as part of this work.",
"cite_spans": [
{
"start": 81,
"end": 100,
"text": "(Lillehaugen, 2016)",
"ref_id": "BIBREF10"
},
{
"start": 202,
"end": 230,
"text": "(Munro & Lopez et al., 1999)",
"ref_id": "BIBREF15"
},
{
"start": 258,
"end": 278,
"text": "(Munro et al., 2021)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Both tools have the potential to support language maintenance efforts. A morphological transducer can be used in various types of computer-assisted language learning software, such as for learning vocabulary (Katinskaia et al., 2018) and complex inflectional systems . FSTs are also used in electronic corpora (Saykhunov et al., 2019) , paradigm generators, 2 text-reading tools, 3 and form-lookup dictionaries . FSTs may be trivially converted to spell checkers (Washington et al., 2021) and can also be used in other types of text-proofing and language-learning tools (e.g., Antonsen, 2012) ; they can further serve as core elements of machine translation systems (Khanna et al., 2021) .",
"cite_spans": [
{
"start": 208,
"end": 233,
"text": "(Katinskaia et al., 2018)",
"ref_id": "BIBREF6"
},
{
"start": 310,
"end": 334,
"text": "(Saykhunov et al., 2019)",
"ref_id": "BIBREF23"
},
{
"start": 380,
"end": 381,
"text": "3",
"ref_id": null
},
{
"start": 463,
"end": 488,
"text": "(Washington et al., 2021)",
"ref_id": "BIBREF25"
},
{
"start": 577,
"end": 592,
"text": "Antonsen, 2012)",
"ref_id": "BIBREF0"
},
{
"start": 666,
"end": 687,
"text": "(Khanna et al., 2021)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Morphological transducers are being developed for languages globally (Khanna et al., 2021) , including for entire language families, such as Turkic (Washington et al., 2021) . Some of these FSTs are developed for languages with large corpora, such as the national languages of Western Europe (Khanna et al., 2021) . One advantage of FSTs is that they can be created for a language without a large quantity of existing text. For example, a morphological trans-ducer has been developed for Zacatl\u00e1n-Ahuacatl\u00e1n-Tepetzintla Nahuatl (Pugh et al., 2021) , a threatened language of Central Mexico with a relatively small corpus of texts. The fact that a morphological transducer can be developed with small corpora creates an entry point especially for threatened languages into the potential benefits of the types of language technology described above.",
"cite_spans": [
{
"start": 69,
"end": 90,
"text": "(Khanna et al., 2021)",
"ref_id": null
},
{
"start": 148,
"end": 173,
"text": "(Washington et al., 2021)",
"ref_id": "BIBREF25"
},
{
"start": 292,
"end": 313,
"text": "(Khanna et al., 2021)",
"ref_id": null
},
{
"start": 528,
"end": 547,
"text": "(Pugh et al., 2021)",
"ref_id": "BIBREF20"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "This paper is structured as follows. Section 2 situates SLQZ and overviews its socio-political context and basic morphological properties. Section 3 describes the morphological transducer and demonstrates several of the challenges which were overcome in its implementation. Section 4 presents a basic evaluation, including na\u00efve coverage and accuracy of orthographic conversion. Section 5 overviews some issues related to licensing of the tools and section 6 concludes.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "San Lucas Quiavin\u00ed Zapotec is spoken by 98% of the population in San Lucas Quiavin\u00ed, Oaxaca, Mexico (DIEGPO, 2015) and by diaspora communities elsewhere in Mexico and the United States, especially the greater Los Angeles area (Lopez and Munro, 1999) , with approximately 3500 total speakers. While children are still acquiring the variety as their first language, it should be considered endangered as the community is shifting to Spanish in more and more contexts (Munro, 2003; P\u00e9rez B\u00e1ez, 2009) .",
"cite_spans": [
{
"start": 226,
"end": 249,
"text": "(Lopez and Munro, 1999)",
"ref_id": "BIBREF15"
},
{
"start": 465,
"end": 478,
"text": "(Munro, 2003;",
"ref_id": "BIBREF16"
},
{
"start": 479,
"end": 496,
"text": "P\u00e9rez B\u00e1ez, 2009)",
"ref_id": "BIBREF22"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "San Lucas Quiavin\u00ed Zapotec",
"sec_num": "2"
},
{
"text": "Western Tlacolula Valley Zapotec encompasses a number of related varieties, with varying degrees of mutual intelligibility. In the present work we focus on the variety of San Lucas Quiavin\u00ed (SLQZ), but we also evaluate the transducer on the variety of San Juan Guelav\u00eda (SJGZ), also classified as Western Tlacolula Valley Zapotec. The two pueblos are separated by no more than 10km, but the two varieties of Zapotec differ in many relevant aspects of their grammar, including tone and phonation contrasts, verbal morphophonology, and pronominal systems.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "San Lucas Quiavin\u00ed Zapotec",
"sec_num": "2"
},
{
"text": "Understanding the morphotactics of SLQZ is essential to developing a morphological transducer. A verb form in SLQZ includes at minimum an aspect marker followed by a verb stem, with very few exceptions. Additionally, a negative-marking enclitic, various other adverbial enclitics (Lee, 2006, 26-27) , and pronominal enclitcs may follow. Nouns generally may be marked as possessed using a prefixwith some suppletive forms and a class of \"essen-tially possessed\" nouns which are always interpreted as possessed. Possessors follow possessed nouns, either as independent noun phrases or as pronominal enclitics. Pronominal enclitics also appear after predicate adjectives.",
"cite_spans": [
{
"start": 280,
"end": 298,
"text": "(Lee, 2006, 26-27)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "San Lucas Quiavin\u00ed Zapotec",
"sec_num": "2"
},
{
"text": "The morphophonology of verb forms in SLQZ is complex. Aspectual prefixes often have multiple realisations. Additionally, there is a large number of verbs whose stems alternate irregularly or are synchronically suppletive depending on aspect, subject, and any following enclitics. Some aspect markers have irregular realisations in these forms. There may also be changes in phonation type before certain enclitics.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "San Lucas Quiavin\u00ed Zapotec",
"sec_num": "2"
},
{
"text": "San Lucas Quiavin\u00ed Zapotec has a very complicated system of tone and phonation with over 23 potential contrasts in a stressed syllable (although see Ch\u00e1vez Pe\u00f3n, 2010 for a different count). Representing all of these contrasts results in an orthography that is complicated. Members of the speech community have directly and indirectly expressed preference for a practical orthography that underrepresents these contrasts. Hence a phonemic orthography (described first in Munro & Lopez et al., 1999) is used in dictionaries and linguistic work, and a simplified orthography (described in Munro et al., 2021) which collapses many phonemic distinctions, is preferred by speakers of the language. Being able to convert the simplified practical orthography to the phonemic orthography would allow linguists and speech scientists to recover the phonemic contrasts from text written in the practical orthography.",
"cite_spans": [
{
"start": 471,
"end": 498,
"text": "Munro & Lopez et al., 1999)",
"ref_id": "BIBREF15"
},
{
"start": 587,
"end": 606,
"text": "Munro et al., 2021)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "San Lucas Quiavin\u00ed Zapotec",
"sec_num": "2"
},
{
"text": "The transducer was implemented manually using the two-level approach (Koskenniemi, 1983) and is designed for use with HFST (Lind\u00e9n et al., 2011) , an open-source toolkit for finite-state morphology. In the two-level approach to morphology, the lexicon and morphotactics of a language are implemented in one finite-state transducer (FST), the morphophonology is implemented in another, and they two are intersected into a single FST with an analysis side and a form side. For the Dizsha transducer described here, both the morphotactics and morphophonology compile from hand-written patterns, lexicons, and rules. The lexd compiler (Swanson and Howell, 2021) was used to implement the morphotactics, and twol (part of HFST) was used to implement the morphophonology.",
"cite_spans": [
{
"start": 69,
"end": 88,
"text": "(Koskenniemi, 1983)",
"ref_id": "BIBREF8"
},
{
"start": 118,
"end": 144,
"text": "HFST (Lind\u00e9n et al., 2011)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Implementation",
"sec_num": "3"
},
{
"text": "The grammatical patterns of SLQZ were implemented in these formalisms by the first author in part while receiving classroom instruction in the language from the second author and based largely on patterns observed in the first volume of a Dizhsa textbook (Munro et al., 2021) . Later the transducer was expanded and revised in consultation with additional volumes of the textbook and other sources cited here, and under the guidance of the second and third authors, the former of whom is a native speaker and teacher of Dizhsa, and the latter of whom is a linguist with expertise in the language.",
"cite_spans": [
{
"start": 255,
"end": 275,
"text": "(Munro et al., 2021)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Implementation",
"sec_num": "3"
},
{
"text": "In section 3.1, the size and shape of the transducer's lexicon is presented. Section 3.2 discusses some design decisions and how some spelling variants were handled. We explain how some aspects of the language's morphotactics (section 3.3) and morphophonology (section 3.4) were implemented. Section 3.5 presents how orthography conversion was implemented.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Implementation",
"sec_num": "3"
},
{
"text": "The lexical entries of the transducer are divided by stem type based on morphological patterning. Table 1 shows the number of stems of various types in the transducer, and the overall number of stems. Several of these categories span multiple lexicons. For example, under \"verbs\" are counted regular verb stems, irregular verb stems (currently spanning two lexd lexicons), and the copula. Additionally, verbs are subcategorised as intransitive (<iv>), transitive (<tv>), and ditransitive (<dtv>). \"Pronouns\" include both bound and free forms, which must be in separate lexicons due to their different morphological distribution.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Lexicon",
"sec_num": "3.1"
},
{
"text": "Despite being the best studied variety of Western Tlacolula Valley Zapotec, many aspects of the grammar of SLQZ are not fully documented or described. Even when the patterns are understood, it is not clear whether particular phenomena are best accounted for through morphology or syntax.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Design decisions",
"sec_num": "3.2"
},
{
"text": "For this reason, in many cases during the construction of the transducer, more than one implementation option seemed reasonable. For example, we chose to analyse verb stems followed by the negative marker \u2329di\u232a~\u2329dy\u232a as an inflected form of the verb stem, as in uny<v><tv><irre><neg> +\u00ebng<prn><pers><p3><prox><pl> for gunydir\u00ebng. We could also have chosen to analyse it as a verb stem followed by an adverbial enclitic, e.g. uny<v><tv><irre>+di<adv>",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Design decisions",
"sec_num": "3.2"
},
{
"text": "Another such decision is the choice to use verb stems as the lemma for all forms of a verb, and in the case of suppletive stems, the stem that patterns with the habitual aspect. Dictionaries for speakers and learners, such as the glossary in Munro et al. (2021) , use the habitual form (prefix+stem) as the headword for entries. The transducer could just as easily use the habitual form as the lemma.",
"cite_spans": [
{
"start": 242,
"end": 261,
"text": "Munro et al. (2021)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "+\u00ebng<prn><pers><p3><prox><pl>.",
"sec_num": null
},
{
"text": "We made similar decisions regarding the lexicon. Some words in SLQZ have common variant pronunciations and corresponding spellings. For example, the word for 'fish' may be spelled bel or beld. In this case the lemma was chosen to be beld, but the generated form was chosen to be bel. The form beld is still analysed to the same lemma. This was implemented by adding the entry to the transducer with both spellings, and including a comment on the analyse-only variant that triggers the compiler to remove that line while creating the generator, but not the analyser. The lines corresponding to these entries are shown in Code Block 1.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "+\u00ebng<prn><pers><p3><prox><pl>.",
"sec_num": null
},
{
"text": "beld:bel behlld:behll # \"fish\" beld:beld behlld:behlld # \"fish\" ! Dir/LR Code Block 1: The entries in the lexd file for the word for 'fish'. All material after the # symbol is ignored by the compiler, but a preprocessing command strips all lines containing Dir/LR before compiling the generator transducer (but not the analyser transducer).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "+\u00ebng<prn><pers><p3><prox><pl>.",
"sec_num": null
},
{
"text": "These analyses reflect our best current understanding of the grammar, but it would be trivial to change the implementation in the future.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "+\u00ebng<prn><pers><p3><prox><pl>.",
"sec_num": null
},
{
"text": "A verb in SLQZ includes an obligatory prefix that signals aspect, optional endings that include a verbal extender (adding politeness) and a negative morpheme, and optional pronominal clitics. This was implemented fairly straightforwardly by defining a general pattern in lexd, shown in Code Block 2.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphotactics",
"sec_num": "3.3"
},
{
"text": "( :Aspect ( V-Stems(1) [<v>:] V-Stems(3): ) V-Extender(1)? Aspect: ) V-Neg(1)? Prn-Bound(1)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphotactics",
"sec_num": "3.3"
},
{
"text": "Code Block 2: The pattern used for regular verbs in the SLQZ transducer. The numbers in parentheses after each element reference \"components\", described in section 3.5. The : character indicates separation of analysis and form. The ? character represents optionality. The parentheses after lexicon names indicate column numbers within lexicons. The parentheses grouping parts of the pattern are not strictly necessary, but speed up compilation due to how matching works (described below).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphotactics",
"sec_num": "3.3"
},
{
"text": "The reason lexd was used instead of HFST's lexc or Lttoolbox's dix formats-the most common choices for implementing a transducer of this type-is because dix is not ideal for agglutinative patterns and lexc requires complicated tricks (flag diacritics or filter transducers) to implement prefixational morphology. The conventional structure of tag-based morphological analyses is a lemma followed by a part of speech tag, followed by any subcategory tags, followed by any grammatical tags. In an example like runy (form) uny<v><tv><hab> (analysis), the analysis presents that uny is the lemma (in this case a verb stem), <v> (verb) is the category of the word, <tv> (transitive) is the subcategory of the word, and <hab> (habitual) is a grammatical property of the form. Thus in a transducer we can define the form-analysis pairs <hab>:r and uny<v><tv>:uny, but if combined in that order, the result would be unconventional <hab>uny<v><tv>:runy.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphotactics",
"sec_num": "3.3"
},
{
"text": "The solution to this is lexicon matching, a feature unique to lexd. For SLQZ, we can create an Aspect lexicon (containing prefixes paired to their analyses, e.g. <hab>:r) and a V-Stems lexicon (which lists regular verbs). In the pattern that combines these lexicons shown in Code Block 2, the lexd compiler keeps track of multiple mentions of a lexicon and matches them. That is, instead of producing forms with all combinations of aspectual prefixes and tags, only the elements of pairs on the same line are used, despite the fact that the elements are referenced at different places in the pattern.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphotactics",
"sec_num": "3.3"
},
{
"text": "Another lexd-specific feature employed is columns within lexicons. In the pattern, columns 1 and 3 of the V-Stems lexicon are referenced. These contain the simple-orthography form of verbs and the subcategory (transitivity) tag, respectively. Some SLQZ verbs have irregular alternations in their stems when combined with perfective aspect prefixes or a first person plural (1PL) subject. This was implemented using filters in lexd, which allows for entries in a given lexicon which are tagged a certain way to be referenced from patterns, to the exclusion of other entries in that lexicon. In this way, separate patterns can be constructed that pull, e.g., (1) only the 1PL stems and pronoun forms, and (2) only the non-1PL stems and pronoun forms.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphotactics",
"sec_num": "3.3"
},
{
"text": "Many of the phonological alternations in SLQZ verb forms are regular. For example, the negative marker is written before a vowel as \u2329dy\u232a, as in queity runydyai / que\u02bcity ruhnydya\u02bcih 'I don't do it' and elsewhere as \u2329di\u232a (simplified orthography) \u2329di\u02bc\u232a (phonemic orthography), as in queity runydi Jwanyi / que\u02bcity ruhnydi\u02bc Jwaanyih 'Juan doesn't do it'.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphophonology",
"sec_num": "3.4"
},
{
"text": "This alternation is implemented by specifying the morpheme with a special character in the morphotactic transducer (lexd), as <neg>:d{I} and <neg>:d{I}' (depending on orthography), and then controlling the alternation of the {I} character using a morphophonology transducer (written in twol).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphophonology",
"sec_num": "3.4"
},
{
"text": "The twol formalism allows for symbol mappings to be restricted based on context. The mappings needed to condition the correct forms of the \u2329di(\u02bc)\u232a/\u2329dy\u232a alternation are presented in Code Block 3. The compiled FST is intersected with the morphotactic transducer to produce correct forms.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphophonology",
"sec_num": "3.4"
},
{
"text": "\"di\u02bc \u2192 dy before vowels: {I}\" %{I%}:y <=> _ (\u02bc:) %>:* :0* :Vow ; \"di\u02bc \u2192 dy before vowels: \u02bc\" \u02bc:0 <=> %{I%}:y _ ;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphophonology",
"sec_num": "3.4"
},
{
"text": "Code Block 3: Morphophonological mapping restrictions specified in the twol formalism to condition the alternation of d{I}(\u02bc) as \u2329dy\u232a before vowels. In other contexts, {I} is realised as \u2329i\u232a. 4 In the transducer's twol file, there are currently 10 characters like {I} defined, and 20 mapping restrictions specified.",
"cite_spans": [
{
"start": 192,
"end": 193,
"text": "4",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Verbal morphophonology",
"sec_num": "3.4"
},
{
"text": "This section outlines both the orthographic support of the transducer and how it is able to be used to convert between orthographies.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "The morphological transducer is compiled into two generators: one for each of the simple and phonemic orthographies. A single analyser is compiled that supports both. This is made possible through a combination of the lexd features of lexicon matching and columns in lexicons, both discussed in section 3.3. For example the phonemicorthography pattern for regular verbs is shown in Code Block 4, and can be compared to the pattern used for simple-orthography regular verbs shown in Code Block 2. The difference between these patterns lies in which column of the lexicons are referenced on the form side. For example, the verb stem lexicon is referenced using",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "V-Stems(1):V-Stems(2) instead of V-Stems(1) (equivalent to V-Stems(1):V-Stems(1)).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "The second column of the V-Stem lexicon (and most lexicons in the transducer, cf. Code Block 1) is the phonemic-orthography form of each stem. The two sides of the lexicon are matched, as opposed to all elements of the first column being paired with all elements of the second column.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "( :Aspect ( V-Stems(1):V-Stems(2) [<v>:] V-Stems(3): ) V-Extender(2)? Aspect: ) V-Neg(2)? Prn-Bound(1):Prn-Bound(2)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "Code Block 4: The pattern used for phonemicorthography regular verbs used in the SLQZ transducer.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "The other crucial part of this approach is control symbols in comments at the end of patterns for each orthography. Specifically, Orth/Simp is added to the end of lines containing simple-orthography patterns and Orth/Dict is added to the end of lines containing phonemic-orthography patterns. Then, as part of the compilation process for the transducer in each orthography, lines containing the control symbols for the other orthography are removed. This ensures that each transducer contains only forms in a single orthography. The respective analysers and generators are compiled from these pared-down lexd files, and the two analysers are unioned, resulting in an analyser that supports both orthographies.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "The simple orthography, as discussed in section 2, collapses many of the distinctions made by the phonemic orthography. Because of this, it is mostly trivial to convert from the phonemic orthography to the simple orthography, but not vice versa.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "For example, a word like xyecwa (simple) / x\ua789y\u00e8e\u02bccwa\u02bc (phonemic) 'my dog' can be converted from phonemic to simple orthography by simply removing the diacritics \u2329\ua789\u232a, \u2329\u3009, and \u2329\u02bc\u232a, and simplifying sequences of repeated vowels. The only other changes needed for most words is the simplification of doubled consonant letters \u2329ll\u232a, \u2329mm\u232a, and \u2329nn\u232a, and the removal of \u2329h\u232a after vowels, e.g. behlld \u2192 beld 'fish'; rille\u02bceh \u2192 rile 'knows how to'. However, as these examples show, conversion in the other direction is non-deterministic.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "To convert between the orthographies, then, two transducers which share an interface are intersected along that interface. Specifically, the analyser in one orthography is intersected with the generator in the other orthography along the analysis side of each. This is possible because the analysis side is the same regardless of the orthography. An example of this method applied to one word is shown in Figure 1 .",
"cite_spans": [],
"ref_spans": [
{
"start": 405,
"end": 413,
"text": "Figure 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Orthography",
"sec_num": "3.5"
},
{
"text": "xyecwa \u2195 becw<n><px>+a<prn><pers><p1><sg> \u2229 becw<n><px>+a<prn><pers><p1><sg> \u2195 x\ua789y\u00e8e\u02bccwa\u02bc (phonemic orthography generator) Figure 1 : Demonstration of the intersection of two transducers to create an orthographic converter. In this example, an analysis in the simplified orthography analyser is matched to an analysis in the phonemic orthography generator, so that when a simplified orthography form is input to the resulting transducer, the corresponding phonemic orthography form is generated.",
"cite_spans": [],
"ref_spans": [
{
"start": 123,
"end": 131,
"text": "Figure 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "(simplified orthography analyser)",
"sec_num": null
},
{
"text": "This approach provides fairly deterministic output, although as discussed in section 4.3, it does not solve the issue of simple-orthography homography.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "(simplified orthography analyser)",
"sec_num": null
},
{
"text": "One additional approach was used to handle orthographic variants, such as any of the apostrophe characters which might be used and the orthography of the Universal Declaration of Human Rights (UDHR) translation, which is like the phonemic or-thography but uses a colon after a vowel to indicate creaky voice, represent by a grave accent over the vowel in the modern version of the phonemic orthography and in the lexd file. A \"spellrelax\" file, containing a series of regular expressions like those shown in Code Block 5, is compiled to an FST and intersected with an analyser. This allows it to accept forms with any of the specified variants used. Code Block 5: Two of the regular expressions contained in the spellrelax file. The first one allows any number of apostrophe characters to be used in place of \u2329\u02bc\u232a, and the second one allows for \u2329a\u232a followed by one of two colon characters to be used in place of \u2329\u00e0\u232a. The .o. symbol conjoins the patterns.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "(simplified orthography analyser)",
"sec_num": null
},
{
"text": "The transducer was evaluated over available texts (4.1) for na\u00efve coverage (4.2) and accuracy of orthographic conversion (4.3).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation",
"sec_num": "4"
},
{
"text": "The transducer was evaluated against a number of available texts, including a number of genres in both the simple and phonemic orthographies.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Texts used for evaluation",
"sec_num": "4.1"
},
{
"text": "The first two parts of the story Blal xte Tiu Pamy\u00ebl (BxTP) are part of Munro et al. (2021) , which is also the source for nearly all of the material in the transducer. A preliminary version of the transducer was evaluated using BxTP parts 1-2, whereafter the transducer was expanded to include unrecognised forms. Hence, BxTP parts 1-2 are treated as development data, and the remaining texts are treated as previously unseen data. Evaluating the transducer over BxTP parts 1-2 also allowed us to observe and correct mismatches between the phonemic and simplified orthographic versions.",
"cite_spans": [
{
"start": 72,
"end": 91,
"text": "Munro et al. (2021)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Texts used for evaluation",
"sec_num": "4.1"
},
{
"text": "A number of poems and stories were also used for evaluation. Those from Tlalocan are in individualised orthographies inspired by the phonemic orthography (Munro, 2014) . There is also a blog post from the Ticha blog entirely in Dizhsa. The Universal Declaration of Human Rights (UDHR) is in an older version of the phonemic orthography, which is easily handled by the transducer due to the addition of some spellrelax mappings.",
"cite_spans": [
{
"start": 154,
"end": 167,
"text": "(Munro, 2014)",
"ref_id": "BIBREF17"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Texts used for evaluation",
"sec_num": "4.1"
},
{
"text": "We also evaluated a translation of the New Testament in SJGZ, a language variety closely related to SLQZ which uses a distinct orthography.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Texts used for evaluation",
"sec_num": "4.1"
},
{
"text": "The complete list of texts is presented in Table 2 , along with na\u00efve coverage results (see section 4.2). The sources for each set of texts are described in footnotes to the table. 5",
"cite_spans": [],
"ref_spans": [
{
"start": 43,
"end": 50,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Texts used for evaluation",
"sec_num": "4.1"
},
{
"text": "Na\u00efve coverage was calculated as the percentage of tokens in a given corpus that received an analysis from the transducer, whether correct or not. Results are shown in Table 2 .",
"cite_spans": [],
"ref_spans": [
{
"start": 168,
"end": 175,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "The results show that the development text has good coverage, at over 90%-higher, not unexpectedly, than coverage over the remaining sources. Unseen texts vary, but average around two thirds coverage, as does the coverage over all available material. This indicates that the transducer has a solid base, but has many opportunities for expansion. It should also be noted that the development text, besides functioning as a graded reader in an introductory textbook for the language, is relatively short, and so lacks a wide range of vocabulary and morphological patterns.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "The lower overall coverage on texts in the phonemic orthography is due primarily to the lack of phonological mappings accounting for all diacritic changes in verb forms, and the homography of the simple orthography. In the simple orthography many words are written the same that are written distinctly in the phonemic orthography. Words that are not in the transducer may receive an incorrect analysis, thus inflating the apparent coverage of texts in the simple orthography.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "The individualised orthographies found in the Tlalocan texts are inspired by, but not the same as, the phonemic orthography, yielding much lower coverage results.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "The translation of the New Testament in the related language variety of SJGZ, totalling 217K tokens, was also evaluated to test whether the SLQZ transducer could be applied to Western Tlacolula 5 The entire set of texts is currently available at https:// github.com/jonorthwash/apertium-zab-corpus. All testing was done on the contents of the transducer repository at revision 0866ec3 and the corpus repository at revision 85fda5c.",
"cite_spans": [
{
"start": 194,
"end": 195,
"text": "5",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "6 Munro et al. (2021) 7 Drawn from Lopez and Lillehaugen (2018) , Lopez and Lillehaugen (2017) Table 2 : Na\u00efve coverage results. BxTP 1-2 was used for development, and the remaining texts were used for testing.",
"cite_spans": [
{
"start": 2,
"end": 21,
"text": "Munro et al. (2021)",
"ref_id": "BIBREF18"
},
{
"start": 35,
"end": 63,
"text": "Lopez and Lillehaugen (2018)",
"ref_id": "BIBREF14"
},
{
"start": 66,
"end": 94,
"text": "Lopez and Lillehaugen (2017)",
"ref_id": "BIBREF13"
}
],
"ref_spans": [
{
"start": 95,
"end": 102,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "Tokens is the number of lexical units according to the transducer, and coverage is the percentage of tokens that received at least one analysis from the transducer.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "Valley Zapotec more broadly. Even with a dedicated spellrelax transducer to account for a number of orthographic differences, the coverage was only a little over 34%. This suggests that perhaps a single transducer for Western Tlacolula Valley Zapotec may not be able to be applied to all varieties.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Na\u00efve coverage",
"sec_num": "4.2"
},
{
"text": "The first four sections of Blal xte Tiu Pamy\u00ebl are available in both the simple and phonemic orthography. To test orthographic conversion, we created two groups of texts, the first group consisting of sections 1 and 2 of BxTP and the second group consisting of sections 3 and 4. The conversion of phonemic to simple orthography is almost entirely deterministic. We set up a simple regular expression (regex) replacement conversion system, which removed diacritics and \u2329h\u232a after vowels and also merged adjacent characters which were identical. The performance of this method provides a baseline measure of similarity between the two texts.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "Performance was measured using Word Error Rate (WER), or the percentage of words that are different between the converted text and the \"gold standard\" of the text in the destination orthography. The results of both the regex-based method and the transducer-based method described in section 3.5 are presented in table 3.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "The performance of the transducer-based approach has a ceiling defined by the level of cover-age and the similarity of the two texts. For example, for phonemic\u2192simple conversion of the first text, it would be impossible to get better (lower) than 8.6% WER, since the text has na\u00efve coverage of 91.4%. None of the words which do not have an analysis in the transducer are able to be converted-although there is a possibility that some of those words would be \"free rides\", or words that are the same in both orthographies. The result of 11.78% WER should be taken in the context of this ceiling.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "In the first group (BxTP 1-2), the simpleto-phonemic conversion performed worse than phonemic-to-simple, despite higher coverage of the source version. This is largely due to homography. While performing disambiguation between available analyses before orthography conversion might improve this result, there are some simpleorthography homographs that may never be possible to accurately decide between (without wider context), such as re, corresponding to both phonemicorthography r\u00e8e 'that' and r\u00e8e\u02bc 'this'.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "The second group of texts (BxTP 3-4) has much lower correspondence between the two orthographies than the first group due to slight differences between the texts, such as words or sentences that seem to be present in one version but absent in the other. That together with the lower coverage over the second group to start with compound for much worse performance.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "While phonemic\u2192simple orthography conversion is deterministic (and hence possible to perform Table 3 : Orthographic conversion accuracy. Tokens is the number of lexical units according to the transducer, coverage is the percentage of tokens that received at least one analysis from the transducer, and WER is word error rate, or the percentage of tokens after orthography conversion that do not correspond to the text in the other orthography.",
"cite_spans": [],
"ref_spans": [
{
"start": 93,
"end": 100,
"text": "Table 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "accurately with a series of regular expressions), sim-ple\u2192phonemic conversion is not, and hence must be done in some other way. These initial experiments in using a lexical approach show that it is a viable method, although it currently suffers from the low overall coverage of the transducer.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Orthographic conversion",
"sec_num": "4.3"
},
{
"text": "We have chosen to license this work under the GNU Affero General Public License (AGPL) because we want it to be available for others to use and build on. This work is also part of a long-term commitment to collaboration with Zapotec communities and community members. The AGPL license allows for uses of our work that would be inconsistent with our commitment to the community. Reciprocity is a defining Zapotec cultural value and practice. Zapotec speakers have shared their knowledge and language in the creation of these resources. Others are allowed to use the tools and in doing so enter into a reciprocal commitment with the Zapotec community that we define in what we call the Guelaguetza clause, shown below:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Licensing",
"sec_num": "5"
},
{
"text": "While licensed under a free/open-source license that permits commercial uses, it is expected that anything created using this resource be made available to the community of San Lucas Quiavin\u00ed free of charge. This is consistent with the community's practice of guelaguetza, a complex system of reciprocity and exchange of goods and labor.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Licensing",
"sec_num": "5"
},
{
"text": "This context reminds us that the more broadly available licenses could use refinements in particular cultural contexts, particularly Indigenous contexts, and that the field should be open to discussions of how culturally specific practices may interact with open source licensing.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Licensing",
"sec_num": "5"
},
{
"text": "This paper has overviewed the development of a morphological transducer and orthography converter for San Lucas Quiavin\u00ed Zapotec.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "An evaluation of the analyser over available texts demonstrates that despite being incomplete, it is fairly robust. Future work to improve the transducer will focus on expanding the lexicon, adding missing morphological patterns, refining the morphophonological patterns, and finding better ways to deal with the nuances of SLQZ verb morphology.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "Text in another variety of Western Tlacolula Valley Zapotec was evaluated using the morphological transducer, and the results suggest that a separate transducer might be needed.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "An evaluation of the orthography converter shows that this method of orthography conversion has potential, but is affected heavily by the coverage of the transducer.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "It is our hope that this resource will be useful to the SLQZ community. In particular, we are excited about the many roles it could play in language maintenance efforts. This work also impacts conversations on language technology for under-resourced languages and open licensing in Indigenous contexts.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6"
},
{
"text": "The tools presented in this paper are available publicly under a free/open-source license https://github.com/ apertium/apertium-zab, and can be used online at https: //beta.apertium.org.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "Such as the prototype at https://apertium.github.io/ apertium-paradigmatrix 3 Such as https://sanit.oahpa.no/read/.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "For more on the twol formalism and its application, see https://github.com/hfst/hfst/wiki/HfstTwolc.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "The authors are grateful to the anonymous reviewers and editors of this volume. We also thank Daniel Swanson for his assistance with lexd. Felipe and Brook express their gratitude to Pamela Munro, who was integral in their understanding of Zapotec grammar. Runybia ra xauzan\u00ebn, ra Bunyza ni za Quiabni, bsan Dizhsa ni zezane\u00ebn nazhi.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgements",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Improving feedback on L2 misspellings -an FST approach",
"authors": [
{
"first": "Lene",
"middle": [],
"last": "Antonsen",
"suffix": ""
}
],
"year": 2012,
"venue": "Proceedings of the SLTC 2012 workshop on NLP for CALL",
"volume": "",
"issue": "",
"pages": "1--10",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lene Antonsen. 2012. Improving feedback on L2 mis- spellings -an FST approach. In Proceedings of the SLTC 2012 workshop on NLP for CALL, pages 1-10.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Generating modular grammar exercises with finite-state transducers",
"authors": [
{
"first": "Lene",
"middle": [],
"last": "Antonsen",
"suffix": ""
},
{
"first": "Ryan",
"middle": [],
"last": "Johnson",
"suffix": ""
},
{
"first": "Trond",
"middle": [],
"last": "Trosterud",
"suffix": ""
},
{
"first": "Heli",
"middle": [],
"last": "Uibo",
"suffix": ""
}
],
"year": 2013,
"venue": "Proceedings of the second workshop on NLP for computer-assisted language learning at NODALIDA 2013",
"volume": "",
"issue": "",
"pages": "27--38",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lene Antonsen, Ryan Johnson, Trond Trosterud, and Heli Uibo. 2013. Generating modular grammar ex- ercises with finite-state transducers. In Proceedings of the second workshop on NLP for computer-assisted language learning at NODALIDA 2013, pages 27-38.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "The interaction of metrical structure, tone, and phonation types in Quiavin\u00ed Zapotec",
"authors": [
{
"first": "Mario",
"middle": [
"E"
],
"last": "Ch\u00e1vez Pe\u00f3n",
"suffix": ""
}
],
"year": 2010,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {
"DOI": [
"10.14288/1.0071253"
]
},
"num": null,
"urls": [],
"raw_text": "Mario E. Ch\u00e1vez Pe\u00f3n. 2010. The interaction of met- rical structure, tone, and phonation types in Quiav- in\u00ed Zapotec. Ph.D. thesis, The University of British Columbia.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Zidgyni zyala rnalaza liu 'Vengo de la luz del amanecer, record\u00e1ndote'. Cuatro poemas y un cuento del zapoteco del Valle",
"authors": [
{
"first": "Mario",
"middle": [
"E"
],
"last": "",
"suffix": ""
}
],
"year": 2009,
"venue": "",
"volume": "16",
"issue": "",
"pages": "17--49",
"other_ids": {
"DOI": [
"10.19130/iifl.tlalocan.2009.199"
]
},
"num": null,
"urls": [],
"raw_text": "Mario E. Ch\u00e1vez Pe\u00f3n and Rom\u00e1n L\u00f3pez Reyes. 2009. Zidgyni zyala rnalaza liu 'Vengo de la luz del amanecer, record\u00e1ndote'. Cuatro poemas y un cuento del zapoteco del Valle. Tlalocan, 16:17-49.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "San Lucas Quiavin\u00ed",
"authors": [
{
"first": "",
"middle": [],
"last": "Diegpo",
"suffix": ""
}
],
"year": 2015,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "DIEGPO. 2015. San Lucas Quiavin\u00ed. libro demogr\u00e1fico.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Using finite state transducers for making efficient reading comprehension dictionaries",
"authors": [
{
"first": "Ryan",
"middle": [],
"last": "Johnson",
"suffix": ""
},
{
"first": "Lene",
"middle": [],
"last": "Antonsen",
"suffix": ""
},
{
"first": "Trond",
"middle": [],
"last": "Trosterud",
"suffix": ""
}
],
"year": 2013,
"venue": "Proceedings of the 19th Nordic Conference of Computational Linguistics (NODALIDA 2013)",
"volume": "",
"issue": "",
"pages": "59--71",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ryan Johnson, Lene Antonsen, and Trond Trosterud. 2013. Using finite state transducers for making ef- ficient reading comprehension dictionaries. In Pro- ceedings of the 19th Nordic Conference of Computa- tional Linguistics (NODALIDA 2013), pages 59-71.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "Revita: a language-learning platform at the intersection of ITS and CALL",
"authors": [
{
"first": "Anisia",
"middle": [],
"last": "Katinskaia",
"suffix": ""
},
{
"first": "Javad",
"middle": [],
"last": "Nouri",
"suffix": ""
},
{
"first": "Roman",
"middle": [],
"last": "Yangarber",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of LREC: 11th International Conference on Language Resources and Evaluation",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Anisia Katinskaia, Javad Nouri, and Roman Yangarber. 2018. Revita: a language-learning platform at the intersection of ITS and CALL. In Proceedings of LREC: 11th International Conference on Language Re- sources and Evaluation, Miyazaki, Japan.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Irene Tang, and H\u00e8ctor Al\u00f2s i Font. 2021. Recent advances in Apertium, a free / open-source rule-based machine translation platform for lowresource languages",
"authors": [
{
"first": "Tanmai",
"middle": [],
"last": "Khanna",
"suffix": ""
},
{
"first": "Jonathan",
"middle": [
"N"
],
"last": "Washington",
"suffix": ""
},
{
"first": "Francis",
"middle": [
"M"
],
"last": "Tyers",
"suffix": ""
},
{
"first": "Sevilay",
"middle": [],
"last": "Bayatl\u0131",
"suffix": ""
},
{
"first": "Daniel",
"middle": [
"G"
],
"last": "Swanson",
"suffix": ""
},
{
"first": "Tommi",
"middle": [
"A"
],
"last": "Pirinen",
"suffix": ""
}
],
"year": null,
"venue": "Machine Translation",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {
"DOI": [
"10.1007/s10590-021-09260-6"
]
},
"num": null,
"urls": [],
"raw_text": "Tanmai Khanna, Jonathan N. Washington, Francis M. Tyers, Sevilay Bayatl\u0131, Daniel G. Swanson, Tommi A. Pirinen, Irene Tang, and H\u00e8ctor Al\u00f2s i Font. 2021. Recent advances in Apertium, a free / open-source rule-based machine translation platform for low- resource languages. Machine Translation.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Two-level Morphology: A General Computational Model for Word-Form Recognition and Production",
"authors": [
{
"first": "Kimmo",
"middle": [],
"last": "Koskenniemi",
"suffix": ""
}
],
"year": 1983,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kimmo Koskenniemi. 1983. Two-level Morphology: A General Computational Model for Word-Form Recog- nition and Production. Ph.D. thesis, University of Helsinki.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Remnant Raising and VSO Clausal Architecture: A Case Study of San Lucas Quiavin\u00ed Zapotec",
"authors": [
{
"first": "Felicia",
"middle": [
"Lee"
],
"last": "",
"suffix": ""
}
],
"year": 2006,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {
"DOI": [
"10.1007/1-4020-4308-2"
]
},
"num": null,
"urls": [],
"raw_text": "Felicia Lee. 2006. Remnant Raising and VSO Clausal Architecture: A Case Study of San Lucas Quiavin\u00ed Za- potec. Springer, Dordrecht.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "Why write in a language that (almost) no one can read? twitter and the development of written literature. Language Documentation and Conservation",
"authors": [
{
"first": "",
"middle": [],
"last": "Brook Danielle Lillehaugen",
"suffix": ""
}
],
"year": 2016,
"venue": "",
"volume": "10",
"issue": "",
"pages": "356--392",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Brook Danielle Lillehaugen. 2016. Why write in a lan- guage that (almost) no one can read? twitter and the development of written literature. Language Docu- mentation and Conservation, 10:356-392.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "HFSTframework for compiling and applying morphologies",
"authors": [
{
"first": "Erik",
"middle": [],
"last": "Krister Lind\u00e9n",
"suffix": ""
},
{
"first": "Sam",
"middle": [],
"last": "Axelson",
"suffix": ""
},
{
"first": "Tommi",
"middle": [
"A"
],
"last": "Hardwick",
"suffix": ""
},
{
"first": "Miikka",
"middle": [],
"last": "Pirinen",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Silfverberg",
"suffix": ""
}
],
"year": 2011,
"venue": "Communications in Computer and Information Science",
"volume": "100",
"issue": "",
"pages": "67--85",
"other_ids": {
"DOI": [
"10.1007/978-3-642-23138-4_5"
]
},
"num": null,
"urls": [],
"raw_text": "Krister Lind\u00e9n, Erik Axelson, Sam Hardwick, Tommi A. Pirinen, and Miikka Silfverberg. 2011. HFST- framework for compiling and applying morphologies. Communications in Computer and Information Sci- ence, 100:67-85.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Liaza chaa / I'm going home",
"authors": [
{
"first": "Felipe",
"middle": [
"H"
],
"last": "Lopez",
"suffix": ""
}
],
"year": 2018,
"venue": "With Brook Danielle Lillehaugen",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Felipe H. Lopez. 2018. Liaza chaa / I'm going home. Latin American Literature Today, 1(7). With Brook Danielle Lillehaugen, translator.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Mam and Guepy: Two Valley Zapotec poems",
"authors": [
{
"first": "Felipe",
"middle": [
"H"
],
"last": "Lopez",
"suffix": ""
},
{
"first": "Brook",
"middle": [
"Danielle"
],
"last": "Lillehaugen",
"suffix": ""
}
],
"year": 2017,
"venue": "Latin America Literary Review",
"volume": "44",
"issue": "88",
"pages": "83--84",
"other_ids": {
"DOI": [
"10.26824/lalr.22"
]
},
"num": null,
"urls": [],
"raw_text": "Felipe H. Lopez and Brook Danielle Lillehaugen. 2017. Mam and Guepy: Two Valley Zapotec poems. Latin America Literary Review, 44(88):83-84.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Seven poems",
"authors": [
{
"first": "Felipe",
"middle": [
"H"
],
"last": "Lopez",
"suffix": ""
},
{
"first": "Brook",
"middle": [
"Danielle"
],
"last": "Lillehaugen",
"suffix": ""
}
],
"year": 2018,
"venue": "Latin American Literature Today",
"volume": "",
"issue": "7",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Felipe H. Lopez and Brook Danielle Lillehaugen, trans- lator. 2018. Seven poems. Latin American Literature Today, 1(7).",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Zapotec immigration: The San Lucas Quiavin\u00ed experience",
"authors": [
{
"first": "Felipe",
"middle": [
"H"
],
"last": "Lopez",
"suffix": ""
},
{
"first": "Pamela",
"middle": [],
"last": "Munro",
"suffix": ""
}
],
"year": 1999,
"venue": "Aztl\u00e1n",
"volume": "24",
"issue": "",
"pages": "129--178",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Felipe H. Lopez and Pamela Munro. 1999. Zapotec im- migration: The San Lucas Quiavin\u00ed experience. Az- tl\u00e1n, 24:129-49.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Preserving the language of the Valley Zapotecs: The orthography question. Presented at Conference on Language and Immigration in France and the United States: Sociolinguistic Perspectives",
"authors": [
{
"first": "Pamela",
"middle": [],
"last": "Munro",
"suffix": ""
}
],
"year": 2003,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Pamela Munro. 2003. Preserving the language of the Valley Zapotecs: The orthography question. Pre- sented at Conference on Language and Immigration in France and the United States: Sociolinguistic Per- spectives.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "Breaking rules for orthography development",
"authors": [
{
"first": "Pamela",
"middle": [],
"last": "Munro",
"suffix": ""
}
],
"year": 2014,
"venue": "Developing orthographies for unwritten languages",
"volume": "",
"issue": "",
"pages": "169--189",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Pamela Munro. 2014. Breaking rules for orthography development. In Michael Cahill and Keren Rice, editors, Developing orthographies for unwritten lan- guages, pages 169-189. SIL International Publica- tions, Dallas.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "Cali Chiu? A Course in Valley Zapotec",
"authors": [
{
"first": "Pamela",
"middle": [],
"last": "Munro",
"suffix": ""
},
{
"first": "Brook",
"middle": [
"Danielle"
],
"last": "Lillehaugen",
"suffix": ""
},
{
"first": "Felipe",
"middle": [
"H"
],
"last": "Lopez",
"suffix": ""
},
{
"first": "Benjamin",
"middle": [],
"last": "Paul",
"suffix": ""
}
],
"year": 2021,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Pamela Munro, Brook Danielle Lillehaugen, and Felipe H. Lopez with Benjamin Paul. 2021. Cali Chiu? A Course in Valley Zapotec, 2nd edition. Haverford Col- lege Libraries Open Educational Resources.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "Lopez with Rodrigo Garcia & Olivia Mendez. 1999. Di'csyonaary X:t\u00e8e'n D\u00eci'zh Sah Sann Luu'c (San Lucas Quiavin\u00ed Zapotec Dictionary / Diccionario Zapoteco de",
"authors": [
{
"first": "Pamela",
"middle": [],
"last": "Munro",
"suffix": ""
},
{
"first": "Felipe",
"middle": [
"H"
],
"last": "",
"suffix": ""
}
],
"year": null,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Pamela Munro and Felipe H. Lopez with Rodrigo Garcia & Olivia Mendez. 1999. Di'csyonaary X:t\u00e8e'n D\u00eci'zh Sah Sann Luu'c (San Lucas Quiavin\u00ed Zapotec Dictio- nary / Diccionario Zapoteco de San Lucas Quiavin\u00ed).",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Towards an open source finite-state morphological analyzer for Zacatl\u00e1n-Ahuacatl\u00e1n-Tepetzintla Nahuatl",
"authors": [
{
"first": "Robert",
"middle": [],
"last": "Pugh",
"suffix": ""
},
{
"first": "Francis",
"middle": [
"M"
],
"last": "Tyers",
"suffix": ""
},
{
"first": "Marivel",
"middle": [
"Huerta"
],
"last": "Mendez",
"suffix": ""
}
],
"year": 2021,
"venue": "Proceedings of the 4th",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {
"DOI": [
"10.33011/computel.v1i.963"
]
},
"num": null,
"urls": [],
"raw_text": "Robert Pugh, Francis M. Tyers, and Marivel Huerta Mendez. 2021. Towards an open source finite-state morphological analyzer for Zacatl\u00e1n-Ahuacatl\u00e1n- Tepetzintla Nahuatl. In Proceedings of the 4th",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "Workshop on Computational Methods for Endangered Languages",
"authors": [],
"year": null,
"venue": "",
"volume": "1",
"issue": "",
"pages": "80--85",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Workshop on Computational Methods for Endangered Languages, volume 1, pages 80-85.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Endangerment of a transnational language: the case of San Lucas Quiavin\u00ed Zapotec",
"authors": [
{
"first": "Gabriela P\u00e9rez",
"middle": [],
"last": "B\u00e1ez",
"suffix": ""
}
],
"year": 2009,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Gabriela P\u00e9rez B\u00e1ez. 2009. Endangerment of a transna- tional language: the case of San Lucas Quiavin\u00ed Za- potec. Ph.D. thesis, State University of New York at Buffalo.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "\u0421\u043b\u043e\u0436\u043d\u043e\u0441\u0442\u0438 \u043f\u0440\u0438 \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0438 \u0442\u0435\u043a\u0441\u0442\u043e\u0432\u043e\u0433\u043e \u043a\u043e\u0440\u043f\u0443\u0441\u0430 \u043e\u0431\u044a\u0435\u043c\u043e\u043c \u0431\u043e\u043b\u0435\u0435 400 \u043c\u043b\u043d \u0442\u043e\u043a\u0435\u043d\u043e\u0432",
"authors": [
{
"first": "M",
"middle": [
"R"
],
"last": "Saykhunov",
"suffix": ""
},
{
"first": "R",
"middle": [
"R"
],
"last": "Khusainov",
"suffix": ""
},
{
"first": "T",
"middle": [
"I"
],
"last": "Ibragimov",
"suffix": ""
}
],
"year": 2019,
"venue": "\u0424\u0438\u043d\u043d\u043e\u0443\u0433\u043e\u0440\u0441\u043a\u0438\u0439 \u043c\u0438\u0440 \u0432 \u043f\u043e\u043b\u0438\u044d\u0442\u043d\u0438\u0447\u043d\u043e\u043c \u043f\u0440\u043e\u0441\u0442\u0440\u0430\u043d\u0441\u0442\u0432\u0435 \u0420\u043e\u0441\u0441\u0438\u0438: \u043a\u0443\u043b\u044c\u0442\u0443\u0440\u043d\u043e\u0435 \u043d\u0430\u0441\u043b\u0435\u0434\u0438\u0435 \u0438 \u043d\u043e\u0432\u044b\u0435 \u0432\u044b\u0437\u043e\u0432\u044b",
"volume": "",
"issue": "",
"pages": "548--554",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M.R. Saykhunov, R.R. Khusainov, and T.I. Ibragimov. 2019. \u0421\u043b\u043e\u0436\u043d\u043e\u0441\u0442\u0438 \u043f\u0440\u0438 \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0438 \u0442\u0435\u043a\u0441\u0442\u043e\u0432\u043e\u0433\u043e \u043a\u043e\u0440- \u043f\u0443\u0441\u0430 \u043e\u0431\u044a\u0435\u043c\u043e\u043c \u0431\u043e\u043b\u0435\u0435 400 \u043c\u043b\u043d \u0442\u043e\u043a\u0435\u043d\u043e\u0432. In \u0424\u0438\u043d\u043d\u043e- \u0443\u0433\u043e\u0440\u0441\u043a\u0438\u0439 \u043c\u0438\u0440 \u0432 \u043f\u043e\u043b\u0438\u044d\u0442\u043d\u0438\u0447\u043d\u043e\u043c \u043f\u0440\u043e\u0441\u0442\u0440\u0430\u043d\u0441\u0442\u0432\u0435 \u0420\u043e\u0441- \u0441\u0438\u0438: \u043a\u0443\u043b\u044c\u0442\u0443\u0440\u043d\u043e\u0435 \u043d\u0430\u0441\u043b\u0435\u0434\u0438\u0435 \u0438 \u043d\u043e\u0432\u044b\u0435 \u0432\u044b\u0437\u043e\u0432\u044b, pages 548-554. UdmFITS UrO RAN.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "Lexd: A finitestate lexicon compiler for non-suffixational morphologies",
"authors": [
{
"first": "Daniel",
"middle": [],
"last": "Swanson",
"suffix": ""
},
{
"first": "Nick",
"middle": [],
"last": "Howell",
"suffix": ""
}
],
"year": 2021,
"venue": "Mika H\u00e4m\u00e4l\u00e4inen, Niko Partanen, and Khalid Alnajjar",
"volume": "",
"issue": "",
"pages": "133--146",
"other_ids": {
"DOI": [
"10.31885/9789515150257.13"
]
},
"num": null,
"urls": [],
"raw_text": "Daniel Swanson and Nick Howell. 2021. Lexd: A finite- state lexicon compiler for non-suffixational morpholo- gies. In Mika H\u00e4m\u00e4l\u00e4inen, Niko Partanen, and Khalid Alnajjar, editors, Multilingual Facilitation, pages 133- 146. Helsingin yliopisto.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Free/open-source technologies for Turkic languages developed in the Apertium project",
"authors": [
{
"first": "Jonathan",
"middle": [
"N"
],
"last": "Washington",
"suffix": ""
},
{
"first": "Ilnar",
"middle": [],
"last": "Salimzianov",
"suffix": ""
},
{
"first": "Francis",
"middle": [
"M"
],
"last": "Tyers",
"suffix": ""
},
{
"first": "Memduh",
"middle": [],
"last": "G\u00f6k\u0131rmak",
"suffix": ""
},
{
"first": "Sardana",
"middle": [],
"last": "Ivanova",
"suffix": ""
},
{
"first": "O\u011fuzhan",
"middle": [],
"last": "Kuyruk\u00e7u",
"suffix": ""
}
],
"year": 2019,
"venue": "Proceedings of the Seventh International Conference on Computer Processing of Turkic Languages",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Jonathan N. Washington, Ilnar Salimzianov, Francis M. Tyers, Memduh G\u00f6k\u0131rmak, Sardana Ivanova, and O\u011fuzhan Kuyruk\u00e7u. 2021. Free/open-source tech- nologies for Turkic languages developed in the Aper- tium project. In Proceedings of the Seventh Interna- tional Conference on Computer Processing of Turkic Languages (TurkLang 2019).",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"type_str": "figure",
"num": null,
"text": "?* [ \u02bc (->) [ %' | %' | %`| %\u00b4| %' | %\u02bb ] ] ?* ] .o. [ ?* ( \u00e0 (->) [ a [ %: | \ua789 ] ] ) ?* ]",
"uris": null
},
"TABREF1": {
"content": "<table/>",
"type_str": "table",
"html": null,
"text": "The size of the transducer's entire lexicon, broken down by individual lexicons, corresponding to lexical category.",
"num": null
},
"TABREF2": {
"content": "<table><tr><td>Use</td><td>Text</td><td>Orthography</td><td colspan=\"2\">Tokens Coverage (%)</td></tr><tr><td colspan=\"3\">development Blal xte Tiu Pamy\u00ebl 1-2 6 Simple</td><td>625</td><td>93.92</td></tr><tr><td/><td/><td>Phonemic</td><td>628</td><td>91.40</td></tr><tr><td>testing</td><td>Blal xte Tiu Pamy\u00ebl 3-7</td><td>Simple</td><td>1532</td><td>73.56</td></tr><tr><td/><td>Blal xte Tiu Pamy\u00ebl 3-4</td><td>Phonemic</td><td>601</td><td>66.89</td></tr><tr><td/><td colspan=\"2\">Felipe H. Lopez poetry 7 Simple</td><td>514</td><td>57.39</td></tr><tr><td/><td colspan=\"2\">Tlalocan poems & story 8 Simple</td><td>635</td><td>57.95</td></tr><tr><td/><td/><td>Individualised</td><td>788</td><td>47.72</td></tr><tr><td/><td>Niny Bac 6</td><td>Simple</td><td>366</td><td>73.77</td></tr><tr><td/><td>Liaza Chaa 9</td><td>Simple</td><td>963</td><td>58.67</td></tr><tr><td/><td colspan=\"2\">Ticha post 2020-07-17 10 Simple</td><td>1026</td><td>60.04</td></tr><tr><td/><td>UDHR (9 articles) 6</td><td>Simple</td><td>433</td><td>69.98</td></tr><tr><td/><td>UDHR (complete) 11</td><td>Phonemic</td><td>1641</td><td>65.63</td></tr><tr><td>total</td><td>all</td><td>mixed</td><td>9934</td><td>67.47</td></tr><tr><td/><td/><td>com/.</td><td/><td/></tr><tr><td/><td/><td colspan=\"3\">8 Ch\u00e1vez Pe\u00f3n and L\u00f3pez Reyes (2009)</td></tr><tr><td/><td/><td>9 Lopez (2018)</td><td/><td/></tr><tr><td/><td/><td colspan=\"3\">10 https://ticha.haverford.edu/updates/</td></tr><tr><td/><td/><td colspan=\"3\">11 https://www.ohchr.org/EN/UDHR/Pages/Language.</td></tr><tr><td/><td/><td>aspx?LangID=ztu1</td><td/><td/></tr></table>",
"type_str": "table",
"html": null,
"text": ", and https://felipehlopez.weebly.",
"num": null
}
}
}
} |