File size: 91,443 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 | {
"paper_id": "P10-1007",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T09:22:30.594037Z"
},
"title": "Correcting errors in speech recognition with articulatory dynamics",
"authors": [
{
"first": "Frank",
"middle": [],
"last": "Rudzicz",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "University of Toronto",
"location": {
"settlement": "Toronto",
"region": "Ontario",
"country": "Canada"
}
},
"email": "frank@cs.toronto.edu"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "We introduce a novel mechanism for incorporating articulatory dynamics into speech recognition with the theory of task dynamics. This system reranks sentencelevel hypotheses by the likelihoods of their hypothetical articulatory realizations which are derived from relationships learned with aligned acoustic/articulatory data. Experiments compare this with two baseline systems, namely an acoustic hidden Markov model and a dynamic Bayes network augmented with discretized representations of the vocal tract. Our system based on task dynamics reduces worderror rates significantly by 10.2% relative to the best baseline models.",
"pdf_parse": {
"paper_id": "P10-1007",
"_pdf_hash": "",
"abstract": [
{
"text": "We introduce a novel mechanism for incorporating articulatory dynamics into speech recognition with the theory of task dynamics. This system reranks sentencelevel hypotheses by the likelihoods of their hypothetical articulatory realizations which are derived from relationships learned with aligned acoustic/articulatory data. Experiments compare this with two baseline systems, namely an acoustic hidden Markov model and a dynamic Bayes network augmented with discretized representations of the vocal tract. Our system based on task dynamics reduces worderror rates significantly by 10.2% relative to the best baseline models.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "Although modern automatic speech recognition (ASR) takes several cues from the biological perception of speech, it rarely models its biological production. The result is that speech is treated as a surface acoustic phenomenon with lexical or phonetic hidden dynamics but without any physical constraints in between. This omission leads to some untenable assumptions. For example, speech is often treated out of convenience as a sequence of discrete, non-overlapping packets, such as phonemes, despite the fact that some major difficulties in ASR, such as co-articulation, are by definition the result of concurrent physiological phenomena (Hardcastle and Hewlett, 1999) .",
"cite_spans": [
{
"start": 639,
"end": 669,
"text": "(Hardcastle and Hewlett, 1999)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Many acoustic ambiguities can be resolved with knowledge of the vocal tract's configuration (O'Shaughnessy, 2000) . For example, the three nasal sonorants, /m/, /n/, and /ng/, are acoustically similar (i.e., they have large concentrations of energy at the same frequencies) but uniquely and reliably involve bilabial closure, tongue-tip elevation, and tongue-dorsum elevation, respectively. Having access to the articulatory goals of the speaker would, in theory, make the identification of linguistic intent almost trivial. Although we don't typically have access to the vocal tract during speech recognition, its configuration can be estimated reasonably well from acoustics alone within adequate models or measurements of the vocal tract (Richmond et al., 2003; Toda et al., 2008) . Evidence that such inversion takes place naturally in humans during speech perception suggests that the discriminability of speech sounds depends powerfully on their production (Liberman and Mattingly, 1985; D'Ausilio et al., 2009) . This paper describes the use of explicit models of physical speech production within recognition systems. Initially, we augment traditional models of ASR with probabilistic relationships between acoustics and articulation learned from appropriate data. This leads to the incorporation of a highlevel, goal-oriented, and control-based theory of speech production within a novel ASR system.",
"cite_spans": [
{
"start": 92,
"end": 113,
"text": "(O'Shaughnessy, 2000)",
"ref_id": "BIBREF18"
},
{
"start": 741,
"end": 764,
"text": "(Richmond et al., 2003;",
"ref_id": "BIBREF19"
},
{
"start": 765,
"end": 783,
"text": "Toda et al., 2008)",
"ref_id": "BIBREF26"
},
{
"start": 963,
"end": 993,
"text": "(Liberman and Mattingly, 1985;",
"ref_id": "BIBREF12"
},
{
"start": 994,
"end": 1017,
"text": "D'Ausilio et al., 2009)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The use of theoretical (phonological) features of the vocal tract has provided some improvement over traditional acoustic ASR systems in phoneme recognition with neural networks (Kirchhoff, 1999; Roweis, 1999) , but there has been very little work in ASR informed by direct measurements of the vocal tract. Recently, Markov et al. (2006) have augmented hidden Markov models with Bayes networks trained to describe articulatory constraints from a small amount of Japanese vocal tract data, resulting in a small phonemeerror reduction. This work has since been expanded upon to inform ASR systems sensitive to physiological speech disorders (Rudzicz, 2009) . Common among previous efforts is an interpretation of speech as a sequence of short, instantaneous observations devoid of long-term dynamics.",
"cite_spans": [
{
"start": 178,
"end": 195,
"text": "(Kirchhoff, 1999;",
"ref_id": "BIBREF11"
},
{
"start": 196,
"end": 209,
"text": "Roweis, 1999)",
"ref_id": "BIBREF20"
},
{
"start": 317,
"end": 337,
"text": "Markov et al. (2006)",
"ref_id": "BIBREF13"
},
{
"start": 639,
"end": 654,
"text": "(Rudzicz, 2009)",
"ref_id": "BIBREF22"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Background and related work",
"sec_num": "2"
},
{
"text": "Articulatory phonology bridges the divide between the physical manifestation of speech and its underlying lexical intentions. Within this discipline, the theory of task dynamics is a combined model of physical articulator motion and the planning of abstract vocal tract configurations (Saltzman, 1986) . This theory introduces the notion that all observed patterns of speech are the result of overlapping gestures, which are abstracted goaloriented reconfigurations of the vocal tract, such as bilabial closure or velar opening (Saltzman and Munhall, 1989) . Each gesture occurs within one of the following tract variables (TVs): velar opening (VEL), lip aperture (LA) and protrusion (LP), tongue tip constriction location (TTCL) and degree (TTCD) 1 , tongue body constriction location (TBCL) and degree (TBCD), lower tooth height (LTH), and glottal vibration (GLO). For example, the syllable pub consists of an onset (/p/), a nucleus (/ah/), and a coda (/b/). Four gestural goals are associated with the onset, namely the shutting of GLO and of VEL, and the closure and release of LA. Similarly, the nucleus of the syllable consists of three goals, namely the relocation of TBCD and TBCL, and the opening of GLO. The presence and extent of these gestural goals are represented by filled rectangles in figure 1. Inter-gestural timings between these goals are specified relative to one another according to human data as described by Nam and Saltzman (2003) . The presence of these discrete goals influences the vocal tract dynamically and continuously as modelled by the following non-homogeneous second-order linear differential equation:",
"cite_spans": [
{
"start": 285,
"end": 301,
"text": "(Saltzman, 1986)",
"ref_id": "BIBREF25"
},
{
"start": 528,
"end": 556,
"text": "(Saltzman and Munhall, 1989)",
"ref_id": "BIBREF24"
},
{
"start": 1433,
"end": 1456,
"text": "Nam and Saltzman (2003)",
"ref_id": "BIBREF17"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory phonology",
"sec_num": "2.1"
},
{
"text": "EQUATION",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [
{
"start": 0,
"end": 8,
"text": "EQUATION",
"ref_id": "EQREF",
"raw_str": "Mz + Bz + K(z \u2212 z * ) = 0.",
"eq_num": "(1)"
}
],
"section": "Articulatory phonology",
"sec_num": "2.1"
},
{
"text": "Here, z is a continuous vector representing the instantaneous positions of the nine tract variables, z * is the target (equilibrium) positions of those variables, and vectors z and z represent the first and second derivatives of z with respect to time (i.e., velocity and acceleration), respectively. The matrices M, B, and K are syllable-specific coefficients describing the inertia, damping, and stiffness, respectively, of the virtual gestures. Generally, this theory assumes that the tract variables are mutually independent, and that the system is critically damped (i.e., the tract variables do not oscillate around their equilibrium positions) (Nam and Saltzman, 2003) . The continuous state, z, of equation (1) is exemplified by black curves in figure 1.",
"cite_spans": [
{
"start": 651,
"end": 675,
"text": "(Nam and Saltzman, 2003)",
"ref_id": "BIBREF17"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory phonology",
"sec_num": "2.1"
},
{
"text": "Tract variables provide the dimensions of an abstract gestural space independent of the physical characteristics of the speaker. In order to complete our articulatory model, however, we require physical data from which to infer these high-level articulatory goals. Electromagnetic articulography (EMA) is a method to measure the motion of the vocal tract during speech. In EMA, the speaker is placed within a low-amplitude electromagnetic field produced within a cube of a known geometry. Tiny sensors within this field induce small electric currents whose energy allows the inference of articulator positions and velocities to within 1 mm of error (Yunusova et al., 2009) . We derive data for the following study from two EMA sources:",
"cite_spans": [
{
"start": 649,
"end": 672,
"text": "(Yunusova et al., 2009)",
"ref_id": "BIBREF28"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory data",
"sec_num": "2.2"
},
{
"text": "\u2022 The University of Edinburgh's MOCHA database, which provides phoneticallybalanced sentences repeated from TIMIT (Zue et al., 1989) uttered by a male and a female speaker (Wrench, 1999) , and",
"cite_spans": [
{
"start": 108,
"end": 132,
"text": "TIMIT (Zue et al., 1989)",
"ref_id": null
},
{
"start": 172,
"end": 186,
"text": "(Wrench, 1999)",
"ref_id": "BIBREF27"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory data",
"sec_num": "2.2"
},
{
"text": "\u2022 The University of Toronto's TORGO database, from which we select sentences repeated from TIMIT from two females and three males (Rudzicz et al., 2008) . (Cerebrally palsied speech, which is the focus of this database, is not included here).",
"cite_spans": [
{
"start": 130,
"end": 152,
"text": "(Rudzicz et al., 2008)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory data",
"sec_num": "2.2"
},
{
"text": "For the following study we use the eight 2D positions common to both databases, namely the upper lip (UL), lower lip (LL), upper incisor (UI), lower incisor (LI), tongue tip (TT), tongue blade (TB), and tongue dorsum (TD). Since these positions are recorded in 3D in TORGO, we project these onto the midsagittal plane. (Additionally, the MOCHA database provides velum (V) data on this plane, and TORGO provides the left and right lip corners (LL and RL) but these are excluded from study except where noted).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory data",
"sec_num": "2.2"
},
{
"text": "All articulatory data is aligned with its associated acoustic data, which is transformed to Melfrequency cepstral coefficients (MFCCs). Since the 2D EMA system in MOCHA and the 3D EMA system in TORGO differ in their recording rates, the length of each MFCC frame in each database must differ in order to properly align acoustics with articulation in time. Therefore, each MFCC frame covers 16 ms in the TORGO database, and 32 ms in MOCHA. Phoneme boundaries are determined automatically in the MOCHA database by forced alignment, and by a speech-language pathologist in the TORGO database.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Articulatory data",
"sec_num": "2.2"
},
{
"text": "We approximate the tract variable space from the physical space of the articulators, in general, through principal component analysis (PCA) on the latter, and subsequent sigmoid normalization on [0, 1]. For example, the LTH tract variable is inferred by calculating the first principal component of the two-dimensional lower incisor (LI) motion in the midsagittal plane, and by normalizing the resulting univariate data through a scaled sigmoid. The VEL variable is inferred similarly from velum (V) EMA data. Tongue tip constriction location and degree (TTCL and TTCD, respectively) are inferred from the 1 st and 2 nd principal components of tongue tip (TT) EMA data, with TBCL and TBCD inferred similarly from tongue body (TB) data. Finally, the glottis (GLO) is inferred by voicing detection on acoustic energy below 150 Hz (O'Shaughnessy, 2000), lip aperture (LA) is the normalized Euclidean distance between the lips, and lip protrusion (LP) is the normalized 2 nd principal component of the midpoint between the lips. All PCA is performed without segmentation of the data. The result is a low-dimensional set of continuous curves describing goal-relevant articulatory variables. Figure 2 , for example, shows the degree of the lip aperture (LA) over time for all instances of the /b/ phoneme in the MOCHA database. The relevant articulatory goal of lip closure is evident.",
"cite_spans": [],
"ref_spans": [
{
"start": 1186,
"end": 1194,
"text": "Figure 2",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "Articulatory data",
"sec_num": "2.2"
},
{
"text": "We now turn to the task of speech recognition. Traditional Bayesian learning is restricted to universal or immutable relationships, and is agnos- tic towards dynamic systems or time-varying relationships. Dynamic Bayes networks (DBNs) are directed acyclic graphs that generalize the powerful stochastic mechanisms of Bayesian representation to temporal sequences. We are free to explicitly provide topological (i.e., dependency) relationships between relevant variables in our models, which can include measurements of tract data.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Baseline systems",
"sec_num": "3"
},
{
"text": "We examine two baseline systems. The first is the standard acoustic hidden Markov model (HMM) augmented with a bigram language model, as shown in figure 3(a). Here, W t \u2192 W t+1 represents word transition probabilities, learned by maximum likelihood estimation, and Ph t \u2192 Ph t+1 represents phoneme transition probabilities whose order is explicitly specified by the relationship W t \u2192 Ph t . Likewise, each phoneme Ph conditions the sub-phoneme state, Q t , whose transition probabilities Q t \u2192 Q t+1 describe the dynamics within phonemes. The variable M t refers to hidden Gaussian indices so that the likelihoods of acoustic observations, O t , are represented by a mixture of 4, 8, 16, or 32 Gaussians for each state and each phoneme. See Murphy (2002) for a further description of this representation. The second baseline model is the articulatory dynamic Bayes network (DBN-A). This augments the standard acoustic HMM by replacing hidden indices, M t , with discrete observations of the vocal tract, K t , as shown in figure 3(b). The pattern of acoustics within each phoneme is dependent on a relatively restricted set of possible articulatory configurations (Roweis, 1999) . To find these discrete positions, we obtain k vectors that best de-scribe the articulatory data according to k-means clustering with the sum-of-squares error function. During training, the DBN variable K t is set explicitly to the index of the mean vector nearest to the current frame of EMA data at time t. In this way, the relationship K t \u2192 O t allows us to learn how discretized articulatory configurations affect acoustics. The training of DBNs involves a specialized version of expectation-maximization, as described in the literature (Murphy, 2002; Ghahramani, 1998) . During inference, variables W t , Ph t , and K t become hidden and we marginalize over their possible values when computing their likelihoods. Bigrams are computed by maximum likelihood on lexical annotations in the training data. ",
"cite_spans": [
{
"start": 742,
"end": 755,
"text": "Murphy (2002)",
"ref_id": "BIBREF15"
},
{
"start": 1165,
"end": 1179,
"text": "(Roweis, 1999)",
"ref_id": "BIBREF20"
},
{
"start": 1723,
"end": 1737,
"text": "(Murphy, 2002;",
"ref_id": "BIBREF15"
},
{
"start": 1738,
"end": 1755,
"text": "Ghahramani, 1998)",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Baseline systems",
"sec_num": "3"
},
{
"text": "M t O t M t+1 O t+1 Q t Ph t Q t+1 Ph t+1 W t W t+1 (a) HMM K t O t K t+1 O t+1 Q t Ph t Q t+1 Ph t+1 W t W t+1 (b) DBN-A",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Baseline systems",
"sec_num": "3"
},
{
"text": "Our first experimental system attempts speech recognition given only articulatory data. The true state of the tract variables at time t \u2212 1 constitutes a 9-dimensional vector, x t\u22121 , of continuous values. Under the task dynamics model of section 2.1, the motions of these tract variables obey critically damped second-order oscillatory relationships. We start with the simplifying assumption of linear dynamics here with allowances for random Gaussian process noise, v t , since articulatory be-haviour is non-deterministic. Moreover, we know that EMA recordings are subject to some error (usually less than 1 mm (Yunusova et al., 2009) ), so the actual observation at time t, y t , will not in general be the true position of the articulators. Assuming that the relationship between y t and x t is also linear, and that the measurement noise, w t , is also Gaussian, then the dynamical articulatory system can be described by",
"cite_spans": [
{
"start": 614,
"end": 637,
"text": "(Yunusova et al., 2009)",
"ref_id": "BIBREF28"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Switching Kalman filter",
"sec_num": "4"
},
{
"text": "EQUATION",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [
{
"start": 0,
"end": 8,
"text": "EQUATION",
"ref_id": "EQREF",
"raw_str": "x t = D t x t\u22121 + v t y t = C t x t + w t .",
"eq_num": "(2)"
}
],
"section": "Switching Kalman filter",
"sec_num": "4"
},
{
"text": "Eqs. 2 form the basis of the Kalman filter which allows us to use EMA measurements directly, rather than quantized abstractions thereof as in the DBN-A model. Obviously, since articulatory dynamics vary significantly for different goals, we replicate eq. (2) for each phoneme and connect these continuous Kalman filters together with discrete conditioning variables for phoneme and word, resulting in the switching Kalman filter (SKF) model. Here, parameters D t and v t are implicit in the relationship x t \u2192 x t+1 , and parameters C t and w t are implicit in x t \u2192 y t . In this model, observation y t is the instantaneous measurements derived from EMA, and x t is their true hidden states. These parameters are trained using expectation-maximization, as described in the literature (Murphy, 1998; Deng et al., 2005 ).",
"cite_spans": [
{
"start": 785,
"end": 799,
"text": "(Murphy, 1998;",
"ref_id": "BIBREF14"
},
{
"start": 800,
"end": 817,
"text": "Deng et al., 2005",
"ref_id": "BIBREF2"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Switching Kalman filter",
"sec_num": "4"
},
{
"text": "Our goal is to integrate task dynamics within an ASR system for continuous sentences called TD-ASR. Our approach is to re-rank an N-best list of sentence hypotheses according to a weighted likelihood of their articulatory realizations. For example, if a word sequence",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition with task dynamics",
"sec_num": "5"
},
{
"text": "W i : w i,1 w i,2 ... w i,m has likelihoods L X (W i ) and L \u039b (W i )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition with task dynamics",
"sec_num": "5"
},
{
"text": "according to purely acoustic and articulatory interpretations of an utterance, respectively, then its overall score would be",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition with task dynamics",
"sec_num": "5"
},
{
"text": "EQUATION",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [
{
"start": 0,
"end": 8,
"text": "EQUATION",
"ref_id": "EQREF",
"raw_str": "L(W i ) = \u03b1L X (W i ) + (1 \u2212 \u03b1)L \u039b (W i )",
"eq_num": "(3)"
}
],
"section": "Recognition with task dynamics",
"sec_num": "5"
},
{
"text": "given a weighting parameter \u03b1 set manually, as in section 6.2. Acoustic likelihoods L X (W i ) are obtained from Viterbi paths through relevant HMMs in the standard fashion.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition with task dynamics",
"sec_num": "5"
},
{
"text": "In order to obtain articulatory likelihoods, L \u039b (W i ), for each word sequence, we first generate articulatory realizations of those sequences according to task dynamics. To this end, we use components from the open-source TADA system (Nam and Goldstein, 2006) , which is a complete implementation of task dynamics. From this toolbox, we use the following components:",
"cite_spans": [
{
"start": 236,
"end": 261,
"text": "(Nam and Goldstein, 2006)",
"ref_id": "BIBREF16"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The TADA component",
"sec_num": "5.1"
},
{
"text": "\u2022 A syllabic dictionary supplemented with the International Speech Lexicon Dictionary (Hasegawa-Johnson and Fleck, 2007) . This breaks word sequences W i into syllable sequences S i consisting of onsets, nuclei, and coda and covers all of MOCHA and TORGO.",
"cite_spans": [
{
"start": 86,
"end": 120,
"text": "(Hasegawa-Johnson and Fleck, 2007)",
"ref_id": "BIBREF9"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The TADA component",
"sec_num": "5.1"
},
{
"text": "\u2022 A syllable-to-gesture lookup table. Given a syllabic sequence, S i , this table provides the gestural goals necessary to produce those syllables. For example, given the syllable pub in figure 1, this table provides the targets for the GLO, VEL, TBCL, and TBCD tract variables, and the parameters for the second-order differential equation, eq. 1, that achieves those goals. These parameters have been empirically tuned by the authors of TADA according to a generic, speakerindependent representation of the vocal tract (Saltzman and Munhall, 1989).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The TADA component",
"sec_num": "5.1"
},
{
"text": "\u2022 A component that produces the continuous tract variable paths that produce an utterance. This component takes into account various physiological aspects of human speech production, including intergestural and interarticulator co-ordination and timing (Nam and Saltzman, 2003; Goldstein and Fowler, 2003) , and the neutral (\"schwa\") forces of the vocal tract (Saltzman and Munhall, 1989) . This component takes a sequence of gestural goals predicted by the segment-to-gesture lookup table, and produces appropriate paths for each tract variable.",
"cite_spans": [
{
"start": 253,
"end": 277,
"text": "(Nam and Saltzman, 2003;",
"ref_id": "BIBREF17"
},
{
"start": 278,
"end": 305,
"text": "Goldstein and Fowler, 2003)",
"ref_id": "BIBREF5"
},
{
"start": 360,
"end": 388,
"text": "(Saltzman and Munhall, 1989)",
"ref_id": "BIBREF24"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The TADA component",
"sec_num": "5.1"
},
{
"text": "The result of the TADA component is a set of N 9-dimensional articulatory paths, TV i , necessary to produce the associated word sequences, W i for i = 1..N. Since task dynamics is a prescriptive model and fully deterministic, TV i sequences are the canonical or default articulatory realizations of the associated sentences. These canonical realizations are independent of our training data, so we transform them in order to more closely resemble the observed articulatory behaviour in our EMA data. Towards this end, we train a switching Kalman filter identical to that in section 4, except the hidden state variable x t is replaced by the observed instantaneous canonical TVs predicted by TADA. In this way we are explicitly learning a relationship between TADA's task dynamics and human data. Since the lengths of these sequences are generally unequal, we align the articulatory behaviour predicted by TADA with training data from MOCHA and TORGO using standard dynamic time warping (Sakoe and Chiba, 1978) . During run-time, the articulatory sequence y t most likely to have been produced by the human data given the canonical sequence TV i is inferred by the Viterbi algorithm through the SKF model with all other variables hidden. The result is a set of articulatory sequences, TV * i , for i = 1..N, that represent the predictions of task dynamics that better resemble our data.",
"cite_spans": [
{
"start": 987,
"end": 1010,
"text": "(Sakoe and Chiba, 1978)",
"ref_id": "BIBREF23"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The TADA component",
"sec_num": "5.1"
},
{
"text": "In order to estimate the articulatory likelihood of an utterance, we need to evaluate each transformed articulatory sequence, TV * i , within probability distributions ranging over all tract variables. These distributions can be inferred using acousticarticulatory inversion. There are a number of approaches to this task, including vector quantization, and expectation-maximization with Gaussian mixtures (Hogden and Valdez, 2001; Toda et al., 2008) . These approaches accurately inferred the xy position of articulators to within 0.41 mm and 2.73 mm. Here, we modify the approach taken by Richmond et al. (2003) , who estimate probability functions over the 2D midsagittal positions of 7 articulators, given acoustics, with a mixturedensity network (MDN). An MDN is essentially a typical discriminative multi-layer neural network whose output consists of the parameters to Gaussian mixtures. Here, each Gaussian mixture describes a probability function over TV positions given the acoustic frame at time t. For example, figure 4 shows an intensity map of the likely values for tongue-tip constriction degree (TTCD) for each frame of acoustics, superimposed with the 'true' trajectory of that TV. Our networks are trained with acoustic and EMA-derived data as described in section 2.2.",
"cite_spans": [
{
"start": 406,
"end": 431,
"text": "(Hogden and Valdez, 2001;",
"ref_id": "BIBREF10"
},
{
"start": 432,
"end": 450,
"text": "Toda et al., 2008)",
"ref_id": "BIBREF26"
},
{
"start": 591,
"end": 613,
"text": "Richmond et al. (2003)",
"ref_id": "BIBREF19"
}
],
"ref_spans": [
{
"start": 1022,
"end": 1030,
"text": "figure 4",
"ref_id": null
}
],
"eq_spans": [],
"section": "Acoustic-articulatory inversion",
"sec_num": "5.2"
},
{
"text": "During recognition of a test utterance, a standard acoustic HMM produces word sequence hypotheses, W i , and associated likelihoods, L(W i ), for i = 1..N. The expected canonical motion of the tract variables, TV i is then produced by task dynamics Figure 4 : Example probability density of tongue tip constriction degree over time, inferred from acoustics. The true trajectory is superimposed as a black curve.",
"cite_spans": [],
"ref_spans": [
{
"start": 249,
"end": 257,
"text": "Figure 4",
"ref_id": null
}
],
"eq_spans": [],
"section": "Recognition by reranking",
"sec_num": "5.3"
},
{
"text": "for each of these word sequences and transformed by an SKF to better match speaker data, giving TV *",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition by reranking",
"sec_num": "5.3"
},
{
"text": "i . The likelihoods of these paths are then evaluated within probability distributions produced by an MDN. The mechanism for producing the articulatory likelihood is shown in figure 5 ",
"cite_spans": [],
"ref_spans": [
{
"start": 175,
"end": 183,
"text": "figure 5",
"ref_id": null
}
],
"eq_spans": [],
"section": "Recognition by reranking",
"sec_num": "5.3"
},
{
"text": ". The overall likelihood, L(W i ) = \u03b1L X (W i ) + (1 \u2212 \u03b1)L \u039b (W i ),",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition by reranking",
"sec_num": "5.3"
},
{
"text": "is then used to produce a final hypothesis list for the given acoustic input.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition by reranking",
"sec_num": "5.3"
},
{
"text": "Experimental data is obtained from two sources, as described in section 2.2. We procure 1200 sentences from Toronto's TORGO database, and 896 from Edinburgh's MOCHA. In total, there are 460 total unique sentence forms, 1092 total unique word forms, and 11065 total words uttered. Except where noted, all experiments randomly split the data into 90% training and 10% testing sets for 5-cross validation. MOCHA and TORGO data are never combined in a single training set due to differing EMA recording rates. In all cases, models are database-dependent (i.e., all TORGO data is conflated, as is all of MOCHA).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "6"
},
{
"text": "For each of our baseline systems, we calculate the phoneme-error-rate (PER) and word-errorrate (WER) after training. The phoneme-errorrate is calculated according to the proportion of frames of speech incorrectly assigned to the proper phoneme. The word-error-rate is calculated as the sum of insertion, deletion, and substitution errors in the highest-ranked hypothesis divided by the total number of words in the correct orthography. The traditional HMM is compared by varying the number of Gaussians used in the modelling of acoustic observations. Similarly, the DBN-A model is compared by varying the number of discrete quantizations of articulatory configurations, as described in section 3. Results are obtained by direct decoding. The average results across both databases, between which there are no significant differences, are shown in table 1. In all cases the DBN-A model outperforms the HMM, which highlights the benefit of explicitly conditioning acoustic observations on articulatory causes.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiments",
"sec_num": "6"
},
{
"text": "In order to evaluate the whole system, we start by evaluating its parts. First, we test how accurately the mixture-density network (MDN) estimates the position of the articulators given only information from the acoustics available during recognition. Table 2 shows the average log likelihood over each tract variable across both databases. These results are consistent with the state-of-the-art (Toda et al., 2008) . In the following experiments, we use MDNs that produce 4 Gaussians. We evaluate how closely transformations to the canonical tract variables predicted by TADA match the data. Namely, we input the known orthography for each test utterance into TADA, obtain the predicted canonical tract variables TV, and transform these according to our trained SKF. The resulting predicted and transformed sequences are aligned with our measurements derived from EMA with dynamic time warping. Finally, we measure the average difference between the observed data and the predicted (canonical and transformed) tract variables. Table 3 shows these differences according to the phonological manner of articulation. In all cases the transformed tract variable motion is more accurate, and significantly so at the 95% confidence level for nasal and retroflex phonemes, and at 99% for fricatives. The practical utility of the transformation component is evaluated in its effect on recognition rates, as described below.",
"cite_spans": [
{
"start": 396,
"end": 415,
"text": "(Toda et al., 2008)",
"ref_id": "BIBREF26"
}
],
"ref_spans": [
{
"start": 252,
"end": 259,
"text": "Table 2",
"ref_id": "TABREF2"
},
{
"start": 1028,
"end": 1035,
"text": "Table 3",
"ref_id": "TABREF4"
}
],
"eq_spans": [],
"section": "Efficacy of TD-ASR components",
"sec_num": "6.1"
},
{
"text": "With the performance of the components of TD-ASR better understood, we combine these and study the resulting composite TD-ASR system. Figure 6 : Word-error-rate according to varying \u03b1, for both TORGO and MOCHA data. Figure 6 shows the WER as a function of \u03b1 with TD-ASR and N = 4 hypotheses per utterance. The effect of \u03b1 is clearly non-monotonic, with articulatory information clearly proving useful. Although systems whose rankings are weighted solely by the articulatory component perform better than the exclusively acoustic systems, the lists available to the former are procured from standard acoustic ASR. Interestingly, the gap between systems trained to the two databases increases as \u03b1 approaches 1.0. Although this gap is not significant, it may be the result of increased inter-speaker articulatory variation in the TORGO database, which includes more than twice as many speakers as MOCHA. Figure 7 shows the WER obtained with TD-ASR given varying-length N-best lists and \u03b1 = 0.7. TD-ASR accuracy at N = 4 is significantly better than both TD-ASR at N = 2 and the baseline approaches of table 1 at the 95% confidence level. However, for N > 4 there is a noticeable and systematic worsening of performance. Figure 7 : Word-error-rate according to varying lengths of N-best hypotheses used, for both TORGO and MOCHA data.",
"cite_spans": [],
"ref_spans": [
{
"start": 134,
"end": 142,
"text": "Figure 6",
"ref_id": null
},
{
"start": 216,
"end": 224,
"text": "Figure 6",
"ref_id": null
},
{
"start": 902,
"end": 910,
"text": "Figure 7",
"ref_id": null
},
{
"start": 1218,
"end": 1226,
"text": "Figure 7",
"ref_id": null
}
],
"eq_spans": [],
"section": "Recognition with TD-ASR",
"sec_num": "6.2"
},
{
"text": "The optimal parameterization of the TD-ASR model results in an average word-error-rate of 8.43%, which represents a 10.3% relative error reduction over the best parameterization of our baseline models. The SKF model of section 4 differs from the HMM and DBN-A baseline models only in its use of continuous (rather than discrete) hidden dynamics and in its articulatory observations. However, its performance is far more variable, and less conclusive. On the MOCHA database the SKF model had an average of 9.54% WER with a standard deviation of 0.73 over 5 trials, and an average of 9.04% WER with a standard deviation of 0.64 over 5 trials on the TORGO database. Despite the presupposed utility of direct articulatory observations, the SKF system does not perform significantly better than the best DBN-A model. Finally, the experiments of tables 6 and 7 are repeated with the canonical tract variables passed untransformed to the probability maps generated by the MDNs. Predictably, resulting articulatory likelihoods L \u039b are less representative and increasing their contribution \u03b1 to the hypothesis reranking does not improve TD-ASR performance significantly, and in some instances worsens it. Although TADA is a useful prescriptive model of generic articulation, its use must be tempered with knowledge of inter-speaker variability.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recognition with TD-ASR",
"sec_num": "6.2"
},
{
"text": "The articulatory medium of speech rarely informs modern speech recognition. We have demonstrated that the use of direct articulatory knowledge can substantially reduce phoneme and word errors in speech recognition, especially if that knowledge is motivated by high-level abstractions of vocal tract behaviour. Task dynamic theory provides a coherent and biologically plausible model of speech production with consequences for phonology (Browman and Goldstein, 1986) , neurolinguistics (Guenther and Perkell, 2004) , and the evolution of speech and language . We have shown that it is also useful within speech recognition.",
"cite_spans": [
{
"start": 436,
"end": 465,
"text": "(Browman and Goldstein, 1986)",
"ref_id": "BIBREF0"
},
{
"start": 485,
"end": 513,
"text": "(Guenther and Perkell, 2004)",
"ref_id": "BIBREF7"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion and conclusions",
"sec_num": "7"
},
{
"text": "We have overcome a conceptual impediment in integrating task dynamics and ASR, which is the former's deterministic nature. This integration is accomplished by stochastically transforming predicted articulatory dynamics and by calculating the likelihoods of these dynamics according to speaker data. However, there are several new avenues for exploration. For example, task dynamics lends itself to more general applications of control theory, including automated self-correction, rhythm, co-ordination, and segmentation (Friedland, 2005) . Other high-level questions also remain, such as whether discrete gestures are the correct biological and practical paradigm, whether a purely continuous representation would be more appropriate, and whether this approach generalizes to other languages.",
"cite_spans": [
{
"start": 520,
"end": 537,
"text": "(Friedland, 2005)",
"ref_id": "BIBREF3"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion and conclusions",
"sec_num": "7"
},
{
"text": "In general, our experiments have revealed very little difference between the use of MOCHA and TORGO EMA data. An ad hoc analysis of some of the errors produced by the TD-ASR system found no particular difference between how systems trained to each of these databases recognized nasal phonemes, although only those trained with MOCHA considered velum motion. Other errors common to both sources of data include phoneme insertion errors, normally vowels, which appear to co-occur with some spurious motion of the tongue between segments, especially for longer N-best lists. Despite the relative slow motion of the articulators relative to acoustics, there remains some intermittent noise.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion and conclusions",
"sec_num": "7"
},
{
"text": "As more articulatory data becomes available and as theories of speech production become more refined, we expect that their combined value to speech recognition will become indispensable.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion and conclusions",
"sec_num": "7"
},
{
"text": "Constriction locations generally refer to the front-back dimension of the vocal tract and constriction degrees generally refer to the top-down dimension.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "This research is funded by the Natural Sciences and Engineering Research Council and the University of Toronto.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgments",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Towards an articulatory phonology",
"authors": [
{
"first": "Catherine",
"middle": [
"P"
],
"last": "Browman",
"suffix": ""
},
{
"first": "Louis",
"middle": [
"M"
],
"last": "Goldstein",
"suffix": ""
}
],
"year": 1986,
"venue": "Phonology Yearbook",
"volume": "3",
"issue": "",
"pages": "219--252",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Catherine P. Browman and Louis M. Goldstein. 1986. To- wards an articulatory phonology. Phonology Yearbook, 3:219-252.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "The motor somatotopy of speech perception",
"authors": [
{
"first": "Alessandro D'",
"middle": [],
"last": "Ausilio",
"suffix": ""
},
{
"first": "Friedemann",
"middle": [],
"last": "Pulvermuller",
"suffix": ""
},
{
"first": "Paola",
"middle": [],
"last": "Salmas",
"suffix": ""
},
{
"first": "Ilaria",
"middle": [],
"last": "Bufalari",
"suffix": ""
},
{
"first": "Chiara",
"middle": [],
"last": "Begliomini",
"suffix": ""
},
{
"first": "Luciano",
"middle": [],
"last": "Fadiga",
"suffix": ""
}
],
"year": 2009,
"venue": "Current Biology",
"volume": "19",
"issue": "5",
"pages": "381--385",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Alessandro D'Ausilio, Friedemann Pulvermuller, Paola Salmas, Ilaria Bufalari, Chiara Begliomini, and Luciano Fadiga. 2009. The motor somatotopy of speech percep- tion. Current Biology, 19(5):381-385, February.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Speech Enhancement Using a Switching Kalman Filter with a Perceptual Post-Filter",
"authors": [
{
"first": "Jianping",
"middle": [],
"last": "Deng",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Bouchard",
"suffix": ""
},
{
"first": "Tet",
"middle": [],
"last": "Yeap",
"suffix": ""
}
],
"year": 2005,
"venue": "Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on",
"volume": "1",
"issue": "",
"pages": "18--23",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Jianping Deng, M. Bouchard, and Tet Yeap. 2005. Speech Enhancement Using a Switching Kalman Filter with a Per- ceptual Post-Filter. In Acoustics, Speech, and Signal Pro- cessing, 2005. Proceedings. (ICASSP '05). IEEE Interna- tional Conference on, volume 1, pages 1121-1124, 18-23,.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Control System Design: An Introduction to State-Space Methods",
"authors": [
{
"first": "Bernard",
"middle": [],
"last": "Friedland",
"suffix": ""
}
],
"year": 2005,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Bernard Friedland. 2005. Control System Design: An Intro- duction to State-Space Methods. Dover.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Learning dynamic Bayesian networks",
"authors": [
{
"first": "Zoubin",
"middle": [],
"last": "Ghahramani",
"suffix": ""
}
],
"year": 1998,
"venue": "Adaptive Processing of Sequences and Data Structures",
"volume": "",
"issue": "",
"pages": "168--197",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zoubin Ghahramani. 1998. Learning dynamic Bayesian net- works. In Adaptive Processing of Sequences and Data Structures, pages 168-197. Springer-Verlag.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Articulatory phonology: a phonology for public language use. Phonetics and Phonology in Language Comprehension and Production: Differences and Similarities",
"authors": [
{
"first": "M",
"middle": [],
"last": "Louis",
"suffix": ""
},
{
"first": "Carol",
"middle": [],
"last": "Goldstein",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Fowler",
"suffix": ""
}
],
"year": 2003,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Louis M. Goldstein and Carol Fowler. 2003. Articulatory phonology: a phonology for public language use. Phonet- ics and Phonology in Language Comprehension and Pro- duction: Differences and Similarities.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "The role of vocal tract gestural action units in understanding the evolution of phonology",
"authors": [
{
"first": "Louis",
"middle": [],
"last": "Goldstein",
"suffix": ""
},
{
"first": "Dani",
"middle": [],
"last": "Byrd",
"suffix": ""
},
{
"first": "Elliot",
"middle": [],
"last": "Saltzman",
"suffix": ""
}
],
"year": 2006,
"venue": "Action to Language via the Mirror Neuron System",
"volume": "",
"issue": "",
"pages": "215--249",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Louis Goldstein, Dani Byrd, and Elliot Saltzman. 2006. The role of vocal tract gestural action units in understanding the evolution of phonology. In M.A. Arib, editor, Action to Language via the Mirror Neuron System, pages 215- 249. Cambridge University Press, Cambridge, UK.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "A neural model of speech production and its application to studies of the role of auditory feedback in speech",
"authors": [
{
"first": "H",
"middle": [],
"last": "Frank",
"suffix": ""
},
{
"first": "Joseph",
"middle": [
"S"
],
"last": "Guenther",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Perkell",
"suffix": ""
}
],
"year": 2004,
"venue": "Speech Motor Control in Normal and Disordered Speech",
"volume": "",
"issue": "",
"pages": "29--49",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Frank H. Guenther and Joseph S. Perkell. 2004. A neu- ral model of speech production and its application to studies of the role of auditory feedback in speech. In Ben Maassen, Raymond Kent, Herman Peters, Pascal Van Lieshout, and Wouter Hulstijn, editors, Speech Motor Control in Normal and Disordered Speech, chapter 4, pages 29-49. Oxford University Press, Oxford.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Coarticulation -Theory, Data, and Techniques",
"authors": [],
"year": 1999,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "William J. Hardcastle and Nigel Hewlett, editors. 1999. Coarticulation -Theory, Data, and Techniques. Cam- bridge University Press.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "International Speech Lexicon Project",
"authors": [
{
"first": "Mark",
"middle": [],
"last": "Hasegawa",
"suffix": ""
},
{
"first": "-Johnson",
"middle": [],
"last": "",
"suffix": ""
},
{
"first": "Margaret",
"middle": [],
"last": "Fleck",
"suffix": ""
}
],
"year": 2007,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Mark Hasegawa-Johnson and Margaret Fleck. 2007. Inter- national Speech Lexicon Project.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "A stochastic articulatory-to-acoustic mapping as a basis for speech recognition",
"authors": [
{
"first": "John",
"middle": [],
"last": "Hogden",
"suffix": ""
},
{
"first": "Patrick",
"middle": [],
"last": "Valdez",
"suffix": ""
}
],
"year": 2001,
"venue": "Proceedings of the 18th IEEE Instrumentation and Measurement Technology Conference",
"volume": "2",
"issue": "",
"pages": "1105--1110",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "John Hogden and Patrick Valdez. 2001. A stochastic articulatory-to-acoustic mapping as a basis for speech recognition. In Proceedings of the 18th IEEE Instrumen- tation and Measurement Technology Conference, 2001. IMTC 2001, volume 2, pages 1105-1110 vol.2.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "Robust Speech Recognition Using Articulatory Information",
"authors": [
{
"first": "Katrin",
"middle": [],
"last": "Kirchhoff",
"suffix": ""
}
],
"year": 1999,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Katrin Kirchhoff. 1999. Robust Speech Recognition Us- ing Articulatory Information. Ph.D. thesis, University of Bielefeld, Germany, July.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "The motor theory of speech perception revised",
"authors": [
{
"first": "Alvin",
"middle": [
"M"
],
"last": "Liberman",
"suffix": ""
},
{
"first": "Ignatius",
"middle": [
"G"
],
"last": "Mattingly",
"suffix": ""
}
],
"year": 1985,
"venue": "Cognition",
"volume": "21",
"issue": "",
"pages": "1--36",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Alvin M. Liberman and Ignatius G. Mattingly. 1985. The motor theory of speech perception revised. Cognition, 21:1-36.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Integration of articulatory and spectrum features based on the hybrid HMM/BN modeling framework",
"authors": [
{
"first": "Konstantin",
"middle": [],
"last": "Markov",
"suffix": ""
},
{
"first": "Jianwu",
"middle": [],
"last": "Dang",
"suffix": ""
},
{
"first": "Satoshi",
"middle": [],
"last": "Nakamura",
"suffix": ""
}
],
"year": 2006,
"venue": "Speech Communication",
"volume": "48",
"issue": "2",
"pages": "161--175",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Konstantin Markov, Jianwu Dang, and Satoshi Nakamura. 2006. Integration of articulatory and spectrum features based on the hybrid HMM/BN modeling framework. Speech Communication, 48(2):161-175, February.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Switching Kalman Filters",
"authors": [
{
"first": "Kevin",
"middle": [
"Patrick"
],
"last": "Murphy",
"suffix": ""
}
],
"year": 1998,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kevin Patrick Murphy. 1998. Switching Kalman Filters. Technical report.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Dynamic Bayesian Networks: Representation, Inference and Learning",
"authors": [
{
"first": "Kevin",
"middle": [
"Patrick"
],
"last": "Murphy",
"suffix": ""
}
],
"year": 2002,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kevin Patrick Murphy. 2002. Dynamic Bayesian Networks: Representation, Inference and Learning. Ph.D. thesis, University of California at Berkeley.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "TADA (TAsk Dynamics Application) manual",
"authors": [
{
"first": "Hosung",
"middle": [],
"last": "Nam",
"suffix": ""
},
{
"first": "Louis",
"middle": [],
"last": "Goldstein",
"suffix": ""
}
],
"year": 2006,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hosung Nam and Louis Goldstein. 2006. TADA (TAsk Dy- namics Application) manual.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "A competitive, coupled oscillator model of syllable structure",
"authors": [
{
"first": "Hosung",
"middle": [],
"last": "Nam",
"suffix": ""
},
{
"first": "Elliot",
"middle": [],
"last": "Saltzman",
"suffix": ""
}
],
"year": 2003,
"venue": "Proceedings of the 15th International Congress of Phonetic Sciences",
"volume": "",
"issue": "",
"pages": "2253--2256",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hosung Nam and Elliot Saltzman. 2003. A competitive, cou- pled oscillator model of syllable structure. In Proceedings of the 15th International Congress of Phonetic Sciences (ICPhS 2003), pages 2253-2256, Barcelona, Spain.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "Speech Communications -Human and Machine",
"authors": [
{
"first": "O'",
"middle": [],
"last": "Douglas",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Shaughnessy",
"suffix": ""
}
],
"year": 2000,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Douglas O'Shaughnessy. 2000. Speech Communications - Human and Machine. IEEE Press, New York, NY, USA.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "Modelling the uncertainty in recovering articulation from acoustics",
"authors": [
{
"first": "Korin",
"middle": [],
"last": "Richmond",
"suffix": ""
},
{
"first": "Simon",
"middle": [],
"last": "King",
"suffix": ""
},
{
"first": "Paul",
"middle": [],
"last": "Taylor",
"suffix": ""
}
],
"year": 2003,
"venue": "Computer Speech and Language",
"volume": "17",
"issue": "",
"pages": "153--172",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Korin Richmond, Simon King, and Paul Taylor. 2003. Modelling the uncertainty in recovering articulation from acoustics. Computer Speech and Language, 17:153-172.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Data Driven Production Models for Speech Processing",
"authors": [
{
"first": "T",
"middle": [],
"last": "Sam",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Roweis",
"suffix": ""
}
],
"year": 1999,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sam T. Roweis. 1999. Data Driven Production Models for Speech Processing. Ph.D. thesis, California Institute of Technology, Pasadena, California.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "Towards a comparative database of dysarthric articulation",
"authors": [
{
"first": "Frank",
"middle": [],
"last": "Rudzicz",
"suffix": ""
},
{
"first": "Graeme",
"middle": [],
"last": "Pascal Van Lieshout",
"suffix": ""
},
{
"first": "Gerald",
"middle": [],
"last": "Hirst",
"suffix": ""
},
{
"first": "Fraser",
"middle": [],
"last": "Penn",
"suffix": ""
},
{
"first": "Talya",
"middle": [],
"last": "Shein",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Wolff",
"suffix": ""
}
],
"year": 2008,
"venue": "Proceedings of the eighth International Seminar on Speech Production (ISSP'08)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Frank Rudzicz, Pascal van Lieshout, Graeme Hirst, Gerald Penn, Fraser Shein, and Talya Wolff. 2008. Towards a comparative database of dysarthric articulation. In Pro- ceedings of the eighth International Seminar on Speech Production (ISSP'08), Strasbourg France, December.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Applying discretized articulatory knowledge to dysarthric speech",
"authors": [
{
"first": "Frank",
"middle": [],
"last": "Rudzicz",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP09)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Frank Rudzicz. 2009. Applying discretized articulatory knowledge to dysarthric speech. In Proceedings of the 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP09), Taipei, Tai- wan, April.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "Dynamic programming algorithm optimization for spoken word recognition",
"authors": [
{
"first": "Hiroaki",
"middle": [],
"last": "Sakoe",
"suffix": ""
},
{
"first": "Seibi",
"middle": [],
"last": "Chiba",
"suffix": ""
}
],
"year": 1978,
"venue": "IEEE Transactions on Acoustics, Speech, and Signal Processing",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hiroaki Sakoe and Seibi Chiba. 1978. Dynamic program- ming algorithm optimization for spoken word recognition. IEEE Transactions on Acoustics, Speech, and Signal Pro- cessing, ASSP-26, February.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "A dynamical approach to gestural patterning in speech production",
"authors": [
{
"first": "L",
"middle": [],
"last": "Elliot",
"suffix": ""
},
{
"first": "Kevin",
"middle": [
"G"
],
"last": "Saltzman",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Munhall",
"suffix": ""
}
],
"year": 1989,
"venue": "Ecological Psychology",
"volume": "1",
"issue": "4",
"pages": "333--382",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Elliot L. Saltzman and Kevin G. Munhall. 1989. A dynam- ical approach to gestural patterning in speech production. Ecological Psychology, 1(4):333-382.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Task dynamic co-ordination of the speech articulators: a preliminary model",
"authors": [
{
"first": "M",
"middle": [],
"last": "Elliot",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Saltzman",
"suffix": ""
}
],
"year": 1986,
"venue": "",
"volume": "",
"issue": "",
"pages": "129--144",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Elliot M. Saltzman, 1986. Task dynamic co-ordination of the speech articulators: a preliminary model, pages 129-144. Springer-Verlag.",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Statistical mapping between articulatory movements and acoustic spectrum using a Gaussian mixture model",
"authors": [
{
"first": "Tomoki",
"middle": [],
"last": "Toda",
"suffix": ""
},
{
"first": "Alan",
"middle": [
"W"
],
"last": "Black",
"suffix": ""
},
{
"first": "Keiichi",
"middle": [],
"last": "Tokuda",
"suffix": ""
}
],
"year": 2008,
"venue": "Speech Communication",
"volume": "50",
"issue": "3",
"pages": "215--227",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Tomoki Toda, Alan W. Black, and Keiichi Tokuda. 2008. Statistical mapping between articulatory movements and acoustic spectrum using a Gaussian mixture model. Speech Communication, 50(3):215-227, March.",
"links": null
},
"BIBREF27": {
"ref_id": "b27",
"title": "The MOCHA-TIMIT articulatory database",
"authors": [
{
"first": "Alan",
"middle": [],
"last": "Wrench",
"suffix": ""
}
],
"year": 1999,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Alan Wrench. 1999. The MOCHA-TIMIT articulatory database, November.",
"links": null
},
"BIBREF28": {
"ref_id": "b28",
"title": "Accuracy Assessment for AG500, Electromagnetic Articulograph",
"authors": [
{
"first": "Yana",
"middle": [],
"last": "Yunusova",
"suffix": ""
},
{
"first": "Jordan",
"middle": [
"R"
],
"last": "Green",
"suffix": ""
},
{
"first": "Antje",
"middle": [],
"last": "Mefferd",
"suffix": ""
}
],
"year": 2009,
"venue": "Journal of Speech, Language, and Hearing Research",
"volume": "52",
"issue": "",
"pages": "547--555",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yana Yunusova, Jordan R. Green, and Antje Mefferd. 2009. Accuracy Assessment for AG500, Electromagnetic Artic- ulograph. Journal of Speech, Language, and Hearing Re- search, 52:547-555, April.",
"links": null
},
"BIBREF29": {
"ref_id": "b29",
"title": "Speech Database Development: TIMIT and Beyond",
"authors": [
{
"first": "Stephanie",
"middle": [],
"last": "Victor Zue",
"suffix": ""
},
{
"first": "James",
"middle": [],
"last": "Seneff",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Glass",
"suffix": ""
}
],
"year": 1989,
"venue": "Proceedings of ESCA Tutorial and Research Workshop on Speech Input/Output Assessment and Speech Databases (SIOA-1989)",
"volume": "2",
"issue": "",
"pages": "35--40",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Victor Zue, Stephanie Seneff, and James Glass. 1989. Speech Database Development: TIMIT and Beyond. In Proceedings of ESCA Tutorial and Research Workshop on Speech Input/Output Assessment and Speech Databases (SIOA-1989), volume 2, pages 35-40, Noordwijkerhout, The Netherlands.",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"uris": null,
"num": null,
"text": "Canonical example pub fromSaltzman and Munhall (1989).",
"type_str": "figure"
},
"FIGREF1": {
"uris": null,
"num": null,
"text": "Lip aperture (LA) over time during all MOCHA instances of /b/.",
"type_str": "figure"
},
"TABREF0": {
"text": "Node W t represents the current word, Ph t is the current phoneme, Q t is that phoneme's dynamic state, O t is the acoustic observation, M t is the Gaussian mixture component, and K t is the discretized articulatory configuration. Filled nodes represent observed variables during training, although only O t is observed during recognition. Square nodes are discrete variables while circular nodes are continuous variables.",
"type_str": "table",
"num": null,
"content": "<table><tr><td>Figure 3: Baseline systems: (a) acoustic hidden</td></tr><tr><td>Markov model and (b) articulatory dynamic Bayes</td></tr><tr><td>network.</td></tr></table>",
"html": null
},
"TABREF2": {
"text": "Average log likelihood of true tract variable positions in test data, under distributions produced by mixture density networks with varying numbers of Gaussians.",
"type_str": "table",
"num": null,
"content": "<table/>",
"html": null
},
"TABREF4": {
"text": "",
"type_str": "table",
"num": null,
"content": "<table><tr><td>: Average difference between predicted</td></tr><tr><td>tract variables and observed data, on [0, 1] scale.</td></tr><tr><td>(*) Nasals are evaluated only with MOCHA data,</td></tr><tr><td>since TORGO data lacks velum measurements.</td></tr></table>",
"html": null
}
}
}
} |