File size: 36,842 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 |
{
"paper_id": "2020",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T04:33:40.326994Z"
},
"title": "Malayalam Speech Corpus: Design and Development for Dravidian Language",
"authors": [
{
"first": "Elizabeth",
"middle": [],
"last": "Sherly",
"suffix": "",
"affiliation": {},
"email": "[email protected]"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "To overpass the disparity between theory and applications in language-related technology in the text as well as speech and several other areas, a well-designed and well-developed corpus is essential. Several problems and issues encountered while developing a corpus, especially for low resource languages. The Malayalam Speech Corpus (MSC) is one of the first open speech corpora for Automatic Speech Recognition (ASR) research to the best of our knowledge. It consists of 250 hours of Agricultural speech data. We are providing a transcription file, lexicon and annotated speech along with the audio segment. It is available in future for public use upon request at \"www.iiitmk.ac.in/vrclc/utilities/ml speechcorpus\". This paper details the development and collection process in the domain of agricultural speech corpora in the Malayalam Language.",
"pdf_parse": {
"paper_id": "2020",
"_pdf_hash": "",
"abstract": [
{
"text": "To overpass the disparity between theory and applications in language-related technology in the text as well as speech and several other areas, a well-designed and well-developed corpus is essential. Several problems and issues encountered while developing a corpus, especially for low resource languages. The Malayalam Speech Corpus (MSC) is one of the first open speech corpora for Automatic Speech Recognition (ASR) research to the best of our knowledge. It consists of 250 hours of Agricultural speech data. We are providing a transcription file, lexicon and annotated speech along with the audio segment. It is available in future for public use upon request at \"www.iiitmk.ac.in/vrclc/utilities/ml speechcorpus\". This paper details the development and collection process in the domain of agricultural speech corpora in the Malayalam Language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "Malayalam is the official language of Kerala, Lakshadweep, and Mahe. From 1330 million people in India, 37 million people speak Malayalam ie; 2.88% of Indians. (Wikipedia contributors, 2020) . Malayalam is the youngest of all languages in the Dravidian family. Four or five decades were taken for Malayalam to emerge from Tamil. The development of Malayalam is greatly influenced by Sanskrit also.",
"cite_spans": [
{
"start": 160,
"end": 190,
"text": "(Wikipedia contributors, 2020)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "In the Automatic Speech Recognition (ASR) area many works are progressing in highly and low-resourced languages. The present speech recognition system has achieved a 'Natural' degree of accuracy mainly in Standard American English (Xiong et al., 2016) . The accurate recognition of speech exists only for highly resourced languages. But it is still lagging for \"non-native\" speakers. To increase the accuracy of such an ASR system the speech data for low-resource language like Malayalam is to be increased.",
"cite_spans": [
{
"start": 231,
"end": 251,
"text": "(Xiong et al., 2016)",
"ref_id": "BIBREF7"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "To encourage the research on speech technology and its related applications in Malayalam, a collection of speech corpus is commissioned and named as Malayalam Speech Corpus (MSC). The corpus consists of the following parts.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "\u2022 200 hours of Narrational Speech named NS and",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "\u2022 50 hours of Interview Speech named IS The raw speech data is collected from \"Kissan Krishideepam\" an agriculture-based program in Malayalam by the Department of Agriculture, Government of Kerala. The NS is created by making a script during the postproduction stage and dubbed with the help of people in different age groups and gender but they are amateur dubbing artists. The speech data is thoughtfully designed -for various applications like code mixed language analysis, Automatic Speech Recognition (ASR) related research, speaker recognition -by considering sociolinguistic variables.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "Interview Speech corpora (NS and IS) collected from native Malayalam speakers. The literature survey of different speech corpora creation is detailed in section 2. Section 3 describes the design and demographics of speech data. The section 4 continues with transcription and section 5 deals with lexicon of the speech data and paper concludes with section 6.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "This paper represents the development of Narrational and",
"sec_num": null
},
{
"text": "Many languages have developed speech corpus and they are open source too. The English read speech corpus is freely available to download for research purposes (Koh et al., 2019) (Panayotov et al., 2015) . Similarly, a database is made available with the collection of TED talks in the English language (Hernandez et al., 2018) . Databases are available for Indian languages on free download and a payment basis also. For the Malayalam language-based emotion recognition, a database is available (Rajan et al., 2019) .",
"cite_spans": [
{
"start": 159,
"end": 177,
"text": "(Koh et al., 2019)",
"ref_id": "BIBREF2"
},
{
"start": 178,
"end": 202,
"text": "(Panayotov et al., 2015)",
"ref_id": "BIBREF3"
},
{
"start": 302,
"end": 326,
"text": "(Hernandez et al., 2018)",
"ref_id": "BIBREF1"
},
{
"start": 495,
"end": 515,
"text": "(Rajan et al., 2019)",
"ref_id": "BIBREF5"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Literature Survey",
"sec_num": "2."
},
{
"text": "The corpus collection of low resourced languages is a good initiative in the area of ASR. One of such work is done on Latvian language (Pinnis et al., 2014) . They created 100 hours of orthographically transcribed audio data and annotated corpus also. In addition to that a four hours of phonetically transcribed audio data is also available. The authors presented the statistics of speech corpus along with criteria for design of speech corpus.",
"cite_spans": [
{
"start": 135,
"end": 156,
"text": "(Pinnis et al., 2014)",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Literature Survey",
"sec_num": "2."
},
{
"text": "South Africa has eleven official languages. An attempt is made for the creation of speech corpora on these under resourced languages (Barnard et al., 2014) . A collection of more than 50 hours of speech in each language is made available. They validated the corpora by building acoustic and language model using KALDI.",
"cite_spans": [
{
"start": 133,
"end": 155,
"text": "(Barnard et al., 2014)",
"ref_id": "BIBREF0"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Literature Survey",
"sec_num": "2."
},
{
"text": "Similarly speech corpora for North-East Indian lowresourced languages is also created (Hernandez et al., 2018) . The authors collected speech and text corpora on Assamese, Bengali and Nepali. They conducted a statisti-cal study of the corpora also.",
"cite_spans": [
{
"start": 86,
"end": 110,
"text": "(Hernandez et al., 2018)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Literature Survey",
"sec_num": "2."
},
{
"text": "A recording studio is setup at our visual media lab with a quiet and sound proof room. A standing microphone is used for recording NS corpora. IS corpora is collected directly from the farmers using recording portable Mic at their place. Hundred speakers are involved in the recording of NS and IS corpora.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Speech Corpora",
"sec_num": "3."
},
{
"text": "The written agricultural script, which is phonetically balanced and phonetically rich (up to triphone model), was given to the speakers to record the Narrational Speech. Scripts were different in content. An example script is provided in Fig:1 . They were given enough time to record the data. If any recording issues happened, after rectification by the recording assistant it was rerecorded. The Narrational Speech is less expensive than Interview Speech because it is difficult to get data for the ASR system. The IS data is collected in a face-to-face interview style. The literacy and the way to communicate information fluently have given less focus. The interviewee with enough experience in his field of cultivation is asked to speak about his cultivation and its features. The interviewer should be preferably a subject expert in the area of cultivation. Both of them are given separate microphones for this purpose.",
"cite_spans": [],
"ref_spans": [
{
"start": 238,
"end": 243,
"text": "Fig:1",
"ref_id": "FIGREF0"
}
],
"eq_spans": [],
"section": "Narrational and Interview Speech Corpora",
"sec_num": "3.1."
},
{
"text": "Few challenges were faced during the recording of the speech corpus. There were lot of background noise like sounds of vehicles, animals, birds, irrigation motor and wind. Another main issue that happened during post production is the difference in pronunciation styles in the Interview Speech corpora collection. This caused difficulty during validation of the corpus. The recording used to extend up to 5-6 hours depending on speakers. The recorded data is then given for post-production to clean unwanted information from that.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Narrational and Interview Speech Corpora",
"sec_num": "3.1."
},
{
"text": "We have set a few criteria for recording the Narrational Speech data.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Speaker Criteria",
"sec_num": "3.2."
},
{
"text": "\u2022 The speakers are at minimum age of 18",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Speaker Criteria",
"sec_num": "3.2."
},
{
"text": "\u2022 They are citizens of India",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Speaker Criteria",
"sec_num": "3.2."
},
{
"text": "\u2022 Speakers are residents of Kerala",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Speaker Criteria",
"sec_num": "3.2."
},
{
"text": "\u2022 The mother tongue of the speaker should be Malayalam without any specific accents",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Speaker Criteria",
"sec_num": "3.2."
},
{
"text": "Speech data is collected with two different microphones for NS and IS. For Narrational Speech, Shure SM58-LC cardioid vocal microphone without cable is used. For IS, we utilized Sennheiser XSW 1-ME2-wireless presentation microphone of range 548-572 MHz Steinberg Nuendo and Pro Tools are used for the audio post-production process.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recording Specifications",
"sec_num": "3.3."
},
{
"text": "The audio is recorded in 48 kHz sampling frequency and 16 bit sampling rate for broadcasting and the same is down sampled to 16 kHz sampling frequency and 16 bit sampling rate for speech-related research purposes. The recordings of speech corpora are saved in WAV files.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Recording Specifications",
"sec_num": "3.3."
},
{
"text": "MSC aims to present a good quality audio recording for speech related research. The NS and IS corpus have both male and female speakers. In NS, the male and female speakers are made up with 75% and 25% respectively. IS have more male speakers than females with 82% and 18% of total speakers. ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Demographics",
"sec_num": "3.4."
},
{
"text": "The NS and IS corpora are transcribed orthographically into Malayalam text. The transcribers are provided with the audio segments that the speaker read. Their task is to transcribe the content of the audio into Malayalam and into phonetic text. A sample of three transcribed data with demographic details is shown below and the annotated speech of first two sentences is depicted in In the first sample a Narrational Speech is detailed. The narrator is about 45 years old and he is describing the details about Palakkad a district in Kerala and a mango estate there. Few sentences are displayed below. ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Transcription",
"sec_num": "4."
},
{
"text": "The pronunciation dictionary, called Lexicon contains a collection of unique 4925 words. The audio collection process is still going on which will increase the lexicon size. The lexicon consists of word and its corresponding phonemic and syllabic representation as in the example shown in ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Lexicon",
"sec_num": "5."
},
{
"text": "Speech is the primary and natural mode of communication than writing. It is possible to extract more linguistic information from speech than text like emotions and accent. Speech related applications are more useful for illiterate and old people. The articulatory and acoustic information can be obtained from a good audio recording environment. One of the important features of speech data is that, there is less interference from a second party compared to textual data.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6."
},
{
"text": "To encourage the academic research in speech related applications, a good number of multilingual and multipurpose speech corpora for Indian languages is required. The responsibility to develop such corpora still lies on the shoulder of the concerned researcher. Also the role of language corpora is very significant to preserve and maintain the linguistic heritage of our country.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6."
},
{
"text": "The release of MSC will be one of the first speech corpora of Malayalam, contributing 200 hours of Narrational Speech and 50 hours of Interview Speech data for public use. The lexicon and annotated speech is also made available with the data. Future work includes creation of corpora related to tourism and entertainment domains and enhancement of quality of speech by building an ASR using KALDI toolkit. The updates on corpus will be accessible through \"www.iiitmk.ac.in/vrclc/utilities/ml speechcorpus\".",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "6."
}
],
"back_matter": [
{
"text": "This research is supported by the Kerala State Council for Science, Technology and Environment (KSCSTE). I thank KSCSTE for funding the project under the Back-to-lab scheme. I also thank Agri team, Indian Institute of Information Technology and Management-Kerala, Kissan Project for collecting the audio data.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgements",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "The nchlt speech corpus of the south african languages",
"authors": [
{
"first": "E",
"middle": [],
"last": "Barnard",
"suffix": ""
},
{
"first": "M",
"middle": [
"H"
],
"last": "Davel",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Van Heerden",
"suffix": ""
},
{
"first": "F",
"middle": [],
"last": "De Wet",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Badenhorst",
"suffix": ""
}
],
"year": 2014,
"venue": "Workshop Spoken Language Technologies for Under-resourced Languages (SLTU)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Barnard, E., Davel, M. H., van Heerden, C., De Wet, F., and Badenhorst, J. (2014). The nchlt speech corpus of the south african languages. In Workshop Spoken Language Technologies for Under-resourced Languages (SLTU).",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Ted-lium 3: twice as much data and corpus repartition for experiments on speaker adaptation",
"authors": [
{
"first": "F",
"middle": [],
"last": "Hernandez",
"suffix": ""
},
{
"first": "V",
"middle": [],
"last": "Nguyen",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Ghannay",
"suffix": ""
},
{
"first": "N",
"middle": [],
"last": "Tomashenko",
"suffix": ""
},
{
"first": "Y",
"middle": [],
"last": "Est\u00e8ve",
"suffix": ""
}
],
"year": 2018,
"venue": "International Conference on Speech and Computer",
"volume": "",
"issue": "",
"pages": "198--208",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hernandez, F., Nguyen, V., Ghannay, S., Tomashenko, N., and Est\u00e8ve, Y. (2018). Ted-lium 3: twice as much data and corpus repartition for experiments on speaker adap- tation. In International Conference on Speech and Com- puter, pages 198-208. Springer.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Building the singapore english national speech corpus",
"authors": [
{
"first": "J",
"middle": [
"X"
],
"last": "Koh",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Mislan",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Khoo",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Ang",
"suffix": ""
},
{
"first": "W",
"middle": [],
"last": "Ang",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Ng",
"suffix": ""
},
{
"first": "Y.-Y",
"middle": [],
"last": "Tan",
"suffix": ""
}
],
"year": 2019,
"venue": "",
"volume": "",
"issue": "",
"pages": "19--22",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Koh, J. X., Mislan, A., Khoo, K., Ang, B., Ang, W., Ng, C., and Tan, Y.-Y. (2019). Building the singapore english national speech corpus. Malay, 20(25.0):19-3.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Librispeech: an asr corpus based on public domain audio books",
"authors": [
{
"first": "V",
"middle": [],
"last": "Panayotov",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Chen",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Povey",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Khudanpur",
"suffix": ""
}
],
"year": 2015,
"venue": "2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)",
"volume": "",
"issue": "",
"pages": "5206--5210",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Panayotov, V., Chen, G., Povey, D., and Khudanpur, S. (2015). Librispeech: an asr corpus based on public do- main audio books. In 2015 IEEE International Con- ference on Acoustics, Speech and Signal Processing (ICASSP), pages 5206-5210. IEEE.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Designing the latvian speech recognition corpus",
"authors": [
{
"first": "M",
"middle": [],
"last": "Pinnis",
"suffix": ""
},
{
"first": "I",
"middle": [],
"last": "Auzina",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Goba",
"suffix": ""
}
],
"year": 2014,
"venue": "LREC",
"volume": "",
"issue": "",
"pages": "1547--1553",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Pinnis, M., Auzina, I., and Goba, K. (2014). Designing the latvian speech recognition corpus. In LREC, pages 1547-1553.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Design and development of a multi-lingual speech corpora (tamar-emodb) for emotion analysis",
"authors": [
{
"first": "R",
"middle": [],
"last": "Rajan",
"suffix": ""
},
{
"first": "U",
"middle": [],
"last": "Haritha",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Sujitha",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "Rejisha",
"suffix": ""
}
],
"year": 2019,
"venue": "Proc. Interspeech 2019",
"volume": "",
"issue": "",
"pages": "3267--3271",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Rajan, R., Haritha, U., Sujitha, A., and Rejisha, T. (2019). Design and development of a multi-lingual speech cor- pora (tamar-emodb) for emotion analysis. Proc. Inter- speech 2019, pages 3267-3271.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "Malayalam -Wikipedia, the free encyclopedia",
"authors": [],
"year": 2020,
"venue": "Wikipedia contributors",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Wikipedia contributors. (2020). Malayalam -Wikipedia, the free encyclopedia. [Online; accessed 21-February- 2020].",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Achieving human parity in conversational speech recognition",
"authors": [
{
"first": "W",
"middle": [],
"last": "Xiong",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Droppo",
"suffix": ""
},
{
"first": "X",
"middle": [],
"last": "Huang",
"suffix": ""
},
{
"first": "F",
"middle": [],
"last": "Seide",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Seltzer",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Stolcke",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Yu",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Zweig",
"suffix": ""
}
],
"year": 2016,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {
"arXiv": [
"arXiv:1610.05256"
]
},
"num": null,
"urls": [],
"raw_text": "Xiong, W., Droppo, J., Huang, X., Seide, F., Seltzer, M., Stolcke, A., Yu, D., and Zweig, G. (2016). Achiev- ing human parity in conversational speech recognition. arXiv preprint arXiv:1610.05256.",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Example of script file for dubbing"
},
"FIGREF1": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Fig 2."
},
"FIGREF2": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "An example of Annotated Speech Corpora Sample 1: Record Entry No : 180220 01 01"
},
"FIGREF4": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Example of the lexicon"
},
"TABREF1": {
"html": null,
"type_str": "table",
"content": "<table/>",
"num": null,
"text": ""
},
"TABREF2": {
"html": null,
"type_str": "table",
"content": "<table><tr><td>Place of Cultivation (District wise)</td><td>IS(%)</td></tr><tr><td>Thiruvananthapuram</td><td>26</td></tr><tr><td>Kollam</td><td>21</td></tr><tr><td>Pathanamthitta</td><td>02</td></tr><tr><td>Ernakulam</td><td>07</td></tr><tr><td>Alappuzha</td><td>08</td></tr><tr><td>Kottayam</td><td>08</td></tr><tr><td>Idukki</td><td>09</td></tr><tr><td>Thrissur</td><td>12</td></tr><tr><td>Wayanad</td><td>03</td></tr><tr><td>Kozhikode</td><td>02</td></tr><tr><td>Kannur</td><td>02</td></tr><tr><td>Total</td><td>100</td></tr></table>",
"num": null,
"text": "and 3 contains the details of the place of cultivation and the type of cultivation in Kerala."
},
"TABREF3": {
"html": null,
"type_str": "table",
"content": "<table><tr><td colspan=\"2\">: Demographic details of speakers by place of culti-</td></tr><tr><td>vation</td><td/></tr><tr><td>Type of Cultivation</td><td>IS (%)</td></tr><tr><td>Animal Husbandry</td><td>10</td></tr><tr><td>Apiculture</td><td>11</td></tr><tr><td>Diary</td><td>16</td></tr><tr><td>Fish and crab farming</td><td>05</td></tr><tr><td>Floriculture</td><td>07</td></tr><tr><td>Fruits and vegetables</td><td>22</td></tr><tr><td>Horticulture</td><td>04</td></tr><tr><td>Mixed farming</td><td>07</td></tr><tr><td>Organic farming</td><td>08</td></tr><tr><td>Poultry</td><td>07</td></tr><tr><td>Terrace farming</td><td>03</td></tr><tr><td>Total</td><td>100</td></tr></table>",
"num": null,
"text": ""
},
"TABREF4": {
"html": null,
"type_str": "table",
"content": "<table/>",
"num": null,
"text": "Demographic details of speakers by type of cultivation"
}
}
}
} |