-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcldfbench_doreco.py
675 lines (637 loc) · 30.5 KB
/
cldfbench_doreco.py
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
"""
This dataset is derived from the DoReCo data as follows:
- DoReCo data is limited to the core data with annotations released under a license without ND (no
derivatives) clause (because we add annotations which goes against this clause).
- Morpheme-aligned data is converted into IGT instances in an ExampleTable.
- Two minor issues with the DoReCo data are fixed, namely a handful of phones in the Evenki corpus
being linked to two words, and phones/words in the Yucatec corpus having a typo in the speaker
reference.
- Links from phones to IPA symbols, and eventually to CLTS sounds are added, based on the
orthography profile in etc/orthography.tsv
Note that the `cldfbench.Dataset` in this module provides functionality to also
- download the ND-licensed data (which is appropriate for analysis, but not for re-destribution)
- download the audio files on which the DoReCo data is based.
To do so, run
- `cldfbench download cldfbench_doreco.py` and answer appropriately when prompted.
- `cldfbench makecldf cldfbench_doreco.py`, passing the location of a clone of cldf-clts/clts when
prompted.
"""
import re
import html
import decimal
import pathlib
import itertools
import subprocess
import collections
import urllib.error
import urllib.parse
import urllib.request
from tqdm import tqdm
import pybtex.database
from pyclts import CLTS
from cldfbench import Dataset as BaseDataset
from cldfbench import CLDFSpec
from clldutils.clilib import confirm
from clldutils.jsonlib import dump, load
from clldutils.markup import add_markdown_text
from clldutils.lgr import ABBRS, PERSONS
from util import nakala
from util import igt
SILENT_PAUSE = '<p:>'
FILLER = '****'
LABEL_PATTERN = re.compile(r'<<(?P<label>fp|fs|pr|fm|sg|bc|id|on|wip|ui)>(?P<content>[^>]+)?>')
LGR_PERSON_AND_ABBR = {p + a for p, a in itertools.product(PERSONS, ABBRS)}
CORPUS_CITATION_FMT = \
"{Creator}. 2022. {Language} DoReCo dataset. In Seifart, Frank, Ludger Paschen and " \
"Matthew Stave (eds.). Language Documentation Reference Corpus (DoReCo) 1.2. Berlin & Lyon: " \
"Leibniz-Zentrum Allgemeine Sprachwissenschaft & laboratoire Dynamique Du Langage " \
"(UMR5596, CNRS & Université Lyon 2). " \
"https://doreco.huma-num.fr/languages/{Glottocode}. DOI:{DOI}"
def is_lgr_abbr(s):
return s in ABBRS or (s in LGR_PERSON_AND_ABBR)
def global_id(glottocode, local_id):
return '{}_{}'.format(glottocode, local_id)
class Dataset(BaseDataset):
dir = pathlib.Path(__file__).parent
id = "doreco"
def cldf_specs(self): # A dataset must declare all CLDF sets it creates.
return CLDFSpec(
dir=self.cldf_dir,
module='Generic',
zipped={'phones.csv', 'words.csv'},
)
def cmd_download(self, args):
self.raw_dir.download(
"https://sharedocs.huma-num.fr/wl/?id=sbLShl5tHQ7J2INRpMaJcotNYWPQioDV&fmode=download", # v1.2
'languages.csv')
self.raw_dir.download(
"https://sharedocs.huma-num.fr/wl/?id=qGGGvFRnpynYiqw8nVvvOQjkl97qhl49&fmode=download", # v1.2
'sources.bib')
with_nd_data = confirm('Include ND data?', default=False)
with_audio_data = confirm('Include audio files?', default=False)
for row in self.raw_dir.read_csv('languages.csv', dicts=True):
if with_nd_data or ('ND' not in row['Annotation license']):
print(row['Glottocode'], row['DOI'])
dep = nakala.Deposit(row['DOI'])
for f in dep.files:
for s in ['_wd.csv', '_ph.csv', '_metadata.csv', '_gloss-abbreviations.csv']:
if f.name.endswith(s):
print(f.name, f.sha1)
self.raw_dir.download(f.url, '{}{}'.format(row['Glottocode'], s))
break
audio = collections.OrderedDict()
for supp in dep.supplements:
for f in supp.files:
if f.mime_type == 'audio/x-wav':
audio[f.name.replace('.wav', '')] = (f.url, f.size)
if with_audio_data:
target = self.dir / 'audio' / row['Glottocode'] / f.name
target.parent.mkdir(parents=True, exist_ok=True)
if not target.exists():
try:
urllib.request.urlretrieve(f.url, target)
except urllib.error.HTTPError as e:
if int(e.code) == 401:
pass
dump(audio, self.raw_dir / '{}_files.json'.format(row['Glottocode']), indent=4)
def cmd_readme(self, args):
# At this point - according to RELEASING.md - .zenodo.json has been written so we can edit
# it, adding the individual corpus citations.
zenodo = load(self.dir / '.zenodo.json')
corpus_citations = [
"Please note that when citing this dataset, it is NOT sufficient to refer to DoReCo as "
"a whole, but the full citation for each individual corpus must be provided, including "
"the name(s) of the creator(s) of each corpus."]
zenodo['description'] += '\n<p>{}</p>\n'.format(corpus_citations[0])
for src in self.cldf_reader().properties['prov:wasDerivedFrom']:
if src.get('dc:title', '').endswith('DoReCo dataset'):
zenodo['description'] += '\n<blockquote>{}</blockquote>\n'.format(
html.escape(src['dc:bibliographicCitation']))
dump(zenodo, self.dir / '.zenodo.json', indent=4)
md = add_markdown_text(
super().cmd_readme(args), '\n'.join(corpus_citations), section='How to cite')
subprocess.check_call([
'cldfbench',
'cldfviz.map',
str(self.cldf_specs().metadata_path),
'--output', str(self.dir / 'map.png'),
'--width', '20',
'--height', '10',
'--format', 'png',
'--with-ocean',
'--language-properties', 'Macroarea',
'--markersize', '15',
'--no-legend',
'--language-labels',
'--padding-top', '3',
'--padding-bottom', '3',
'--pacific-centered'])
return add_markdown_text(
md, """
See [USAGE](USAGE.md) for information how the dataset can be analyszed.
""", section="Description")
def iter_rows(self, pattern):
mismatch = set()
for p in sorted(self.raw_dir.glob(pattern), key=lambda pp: pp.name):
# What to do if there are tab-delimited files? Sniff!
delimiter = ','
with p.open(encoding='utf8') as f:
if '\t' in f.readline(): # Even gloss abbreviations come in a tab-delimited file.
assert 'even' in p.stem
delimiter = '\t'
for row in self.raw_dir.read_csv(p.name, dicts=True, delimiter=delimiter):
# Catch the Beja issue in v1.2, where the file for a different language was
# packaged in the deposit for Beja:
if row.get('lang') and row.get('lang') != p.name.partition('_')[0]:
if p.name.partition('_')[0] not in mismatch:
print('Glottocode mismatch: {}'.format(p))
mismatch.add(p.name.partition('_')[0])
# raise ValueError(p)
row.setdefault('Glottocode', p.name.partition('_')[0])
if '_wd' in pattern or ('_ph' in pattern):
# doreco-mb-algn and mc-zero col missing in some files of _ph
row.setdefault('doreco-mb-algn', '')
row.setdefault('mc-zero', '')
if '_wd' in pattern:
for k in ['ft', 'tx']:
if k in row:
row[k] = igt.fix_text(row[k], k, row['Glottocode'])
yield row
def cmd_makecldf(self, args):
clts_data = pathlib.Path('cldf-clts-clts-6dc73af')
if not clts_data.exists():
clts_data = pathlib.Path(input('Path to clts data: '))
clts = CLTS(clts_data)
xsampa_to_bipa = collections.OrderedDict()
for row in self.etc_dir.read_csv('orthography.tsv', dicts=True, delimiter='\t'):
bipa = clts.bipa[row['IPA']] if row['IPA'] else None
if bipa and bipa.type != 'unknownsound':
xsampa_to_bipa[row['Grapheme']] = bipa
self.create_schema(args.writer.cldf)
args.writer.cldf.add_sources(pybtex.database.parse_string(
self.raw_dir.joinpath('sources.bib').read_text(encoding='utf8'),
bib_format='bibtex',
))
known, i = {}, 0
for xsampa, bipa in xsampa_to_bipa.items():
if bipa.s not in known:
i += 1
args.writer.objects['ParameterTable'].append(dict(
ID=str(i),
Name=bipa.s,
CLTS_ID=bipa.name.replace(' ', '_'),
))
known[bipa.s] = str(i)
xsampa_to_bipa[xsampa] = known[bipa.s]
args.log.info("added sources")
for row in self.raw_dir.read_csv('languages.csv', dicts=True):
if not self.raw_dir.joinpath('{}_metadata.csv'.format(row['Glottocode'])).exists():
continue
args.writer.cldf.add_provenance(wasDerivedFrom={
"rdf:about": "https://github.com/cldf-datasets/doreco/",
"rdf:type": "prov:Entity",
"dc:title": "{} DoReCo dataset".format(row['Language']),
"dc:bibliographicCitation": CORPUS_CITATION_FMT.format(**row)
})
args.writer.objects["LanguageTable"].append({
"ID": row["Glottocode"],
"Name": row["Language"],
"Glottocode": row["Glottocode"],
"Family": row["Family"],
"Latitude": row["Latitude"],
"Longitude": row["Longitude"],
"Macroarea": row["Area"],
"Source": "doreco-" + row["Glottocode"],
})
args.writer.objects["ContributionTable"].append({
"ID": row["Glottocode"],
"Name": "{} DoReCo dataset".format(row["Language"]),
"Citation": CORPUS_CITATION_FMT.format(**row),
"Contributor": row["Creator"],
"Archive": row["Archive"],
"Archive_link": row["Archive_link"],
"AnnotationLicense": row["Annotation license"],
"AudioLicense": row["Audio license"],
"DOI": row["DOI"]
})
args.log.info("added languages and contributions")
speakers = set()
filemd = collections.defaultdict(dict)
for p in self.raw_dir.glob('*_files.json'):
filemd[p.stem.split('_')[0]] = load(p)
for i, row in enumerate(self.iter_rows('*_metadata.csv'), start=1):
if row['extended'] == 'yes':
continue
fid = 'doreco_{}'.format(global_id(row['Glottocode'], row["name"]))
if fid in filemd[row['Glottocode']]:
args.writer.objects["MediaTable"].append({
"ID": fid,
'Name': '{}.wav'.format(row['name']),
"rec_date": row["rec_date"],
"rec_date_assignment_certain": row["rec_date_c"],
"genre": row["genre"],
"genre_stim": row["genre_stim"],
"gloss": row["gloss"],
"transl": row["transl"].split('/'),
"sound_quality": row["sound_quality"],
"background_noise": row["background_noise"],
"Glottocode": row["Glottocode"],
"Corpus_ID": row["Glottocode"],
"Download_URL": filemd[row['Glottocode']][fid][0],
'Media_Type': 'audio/x-wav',
})
age_c = row['spk_age_c'].split('/') if '/' in row['spk_age_c'] else row['spk_age_c']
for i, (code, age, sex) in enumerate(
itertools.zip_longest(
row['spk_code'].split('/'),
row['spk_age'].split('/'),
row['spk_sex'].split('/'))):
assert code and age and sex, '{} {} {} {}'.format(row['Glottocode'], code, age, sex)
code = global_id(row['Glottocode'], code)
if code not in speakers:
args.writer.objects['speakers.csv'].append(dict(
ID=code,
age=None if age == 'na' else int(age),
sex=sex,
age_assignment_certain=age_c[i] if isinstance(age_c, list) else age_c,
Language_ID=row['Glottocode'],
))
speakers.add(code)
for i, row in enumerate(self.iter_rows('*_gloss-abbreviations.csv'), start=1):
args.writer.objects["glosses.csv"].append({
"ID": global_id(row['Glottocode'], str(i)),
"Gloss": row["Gloss"],
"LGR": is_lgr_abbr(row["LGR"]),
"Meaning": row["Meaning"],
"Glottocode": row["Glottocode"]
})
wd_intervals = {} # We store start and end of words - as specified by contained phones.
uid = 0 # We are adding utterance IDs.
gc = None
for wid, rows in tqdm(itertools.groupby(self.iter_rows('*_ph.csv'), lambda r: r['wd_ID']), desc='phones'):
i, core, row, global_wid = 0, True, None, None
while core:
try:
row = next(rows)
except StopIteration: # row is now the last phone in the word.
wd_intervals['{}_{}'.format(gc, wid)][1] = decimal.Decimal(row["end"])
break
core = row['core_extended'] != 'extended'
if not core:
break
start, end = decimal.Decimal(row["start"]), decimal.Decimal(row["end"])
if i == 0: # The first phone in the word.
if row['Glottocode'] != gc:
# A new corpus, make sure we are not conflating utterance.
# FIXME: Should be done per file!
uid += 1
gc = row['Glottocode']
if wid.split()[0] != wid:
# Known problem of the Evenki corpus, see
# https://github.com/DoReCo/doreco/issues/13
assert gc == 'even1259'
wid = wid.split()[-1]
global_wid = global_id(gc, wid)
wd_intervals[global_wid] = [start, None]
speaker = global_id(gc, row['speaker'])
if speaker.startswith('yuca1254_0'):
# Known problem of the Yucatec corpus, see
# https://github.com/DoReCo/doreco/issues/5#issuecomment-1490180631
speaker = speaker.replace('0', '')
assert speaker in speakers, 'Unknown speaker: {}'.format(speaker)
else:
assert start >= args.writer.objects["phones.csv"][-1]['end']
if row['ph'] == SILENT_PAUSE: # Silent pauses delimit utterances.
uid += 1
args.writer.objects["phones.csv"].append({
"ph_ID": gc + "_" + row["ph_ID"],
"ph": row["ph"],
"IPA": xsampa_to_bipa[row['ph']] if row['ph'] in xsampa_to_bipa else None,
"start": start,
"end": end,
"duration": end - start,
"wd_ID": global_wid,
'u_ID': None if row['ph'] == SILENT_PAUSE else str(uid),
'Token_Type': 'pause' if row['ph'] == SILENT_PAUSE else (
'label' if row['ph'].startswith('<<') else 'xsampa'),
})
i += 1
eids = collections.defaultdict(int)
for (f, tx, ft), rows in tqdm(itertools.groupby(self.iter_rows('*_wd.csv'), lambda r: (r['file'], r['tx'], r['ft'])), desc='words'):
rows = list(rows)
eid = None
# Create an entry in ExampleTable if tx not in ['', None, '****', '<p:>']
if tx and ft and ft not in {FILLER, SILENT_PAUSE}:
ex = igt.igt(
rows,
tx,
ft,
eids,
rows[0]['file'] if rows[0]['core_extended'] != 'extended' and rows[0]['file'] in filemd[rows[0]['Glottocode']] else None)
if ex:
args.writer.objects['ExampleTable'].append(ex)
eid = ex['ID']
for row in rows:
gc = row['Glottocode']
wid = global_id(gc, row['wd_ID'])
sid = None
start, end = decimal.Decimal(row["start"]), decimal.Decimal(row["end"])
if wid in wd_intervals:
ps, pe = wd_intervals[wid]
assert start <= ps and pe <= end, 'Conflicting time alignment of wd and ph.'
sid = global_id(gc, row["speaker"])
del wd_intervals[wid]
core = row['core_extended'] != 'extended'
args.writer.objects["words.csv"].append({
"Language_ID": gc,
"File_ID": row["file"] if core and row['file'] in filemd[gc] else None,
"core": core,
# Only speakers for core words are normalized.
"Speaker_ID": sid,
"Example_ID": eid,
"wd_ID": wid,
"wd": row["wd"],
"start": start,
"end": end,
"duration": end - start,
"ref": row["ref"],
"tx": row["tx"],
"ft": row["ft"],
"mb": row["mb"].split(),
# FIXME: add ps and gl to ExampleTable!
"ps": row["ps"].split(),
"gl": row["gl"].split(),
})
assert not wd_intervals, '{} missing wd_IDs linked from phones!'.format(len(wd_intervals))
def create_schema(self, cldf):
t = cldf.add_component(
'MediaTable',
{
'name': 'rec_date',
'datatype': {'base': 'string', 'format': '[0-9]{4}(-[0-9]{2})?(-[0-9]{2})?'},
'dc:description': 'Date of recording. See also rec_date_assignment_certain.',
},
{
'name': 'rec_date_assignment_certain',
'datatype': {'base': 'string', 'format': 'certain|approximate'}},
{
'name': 'genre',
'datatype': {'base': 'string', 'format': 'traditional narrative|personal narrative|conversation|stimulus retelling|procedural|procedural/conversation'}},
{
'name': 'genre_stim',
'null': ['na'],
'datatype': 'string'},
{
'name': 'gloss',
'dc:description': 'Information on whether/how the audio has been annotated with glosses.',
'datatype': 'string',
},
{
'name': 'transl',
'dc:description': 'Information on meta languages for which the annotations contain translations.',
'separator': '/',
'null': ['na'],
'datatype': 'string',
},
{
'name': 'sound_quality',
'datatype': {'base': 'string', 'format': 'good|medium|bad|middle'}},
{
'name': 'background_noise',
'null': ['na'],
'datatype': {'base': 'string', 'format': 'punctual|none|constant|medium'}},
{
'name': 'Corpus_ID',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#contributionReference',
'datatype': 'string',
},
{
'name': 'Glottocode',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#languageReference',
'datatype': 'string',
}
)
t.common_props['dc:description'] = \
"This table lists audio files which have been transcribed/annotated for DoReCo corpora. " \
"Note that only downloadable files are listed."
cldf.remove_columns('MediaTable', 'Description', 'Path_In_Zip')
t = cldf.add_component(
'ExampleTable',
{
'name': 'File_ID',
'dc:description': 'Link to the audio file to which start and end markers pertain.',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#mediaReference',
'datatype': 'string',
},
{
'name': 'start',
'dc:description': 'Start of the word in the linked sound file in (floating point) seconds.',
'datatype': 'decimal',
},
{
'name': 'end',
'dc:description': 'End of the word in the linked sound file in (floating point) seconds.',
'datatype': 'decimal',
},
{
'name': 'duration',
'dc:description': 'Duration of the word in the linked sound file in (floating point) seconds.',
'datatype': 'decimal',
},
)
t.common_props['dc:description'] = \
"Some corpora in DoReCo contain data annotated with glosses. Such glossed data is " \
"extracted into a CLDF ExampleTable."
cldf['ExampleTable', 'Analyzed_Word'].null = []
cldf['ExampleTable', 'Gloss'].null = []
cldf['ExampleTable', 'Analyzed_Word'].separator = '\t'
cldf['ExampleTable', 'Gloss'].separator = '\t'
cldf.add_component(
'LanguageTable',
{'name': 'Source', 'datatype': 'string', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#source'},
{'name': 'Family', 'datatype': 'string'}
)
t = cldf.add_component(
'ContributionTable',
{'name': 'Archive', 'datatype': 'string', 'null': ['na']},
{'name': 'Archive_link', 'datatype': 'string', 'null': ['na']},
{'name': 'AnnotationLicense', 'datatype': {'base': 'string', 'format': 'CC BY|CC BY-NC-SA|CC BY-NC|CC BY-NC-ND'}},
{'name': 'AudioLicense', 'datatype': 'string'},
{'name': 'DOI', 'datatype': 'string'},
)
t.common_props['dc:description'] = \
"Each DoReCo language corpus is listed as separate contribution in this table."
t = cldf.add_component(
'ParameterTable',
{'name': 'CLTS_ID', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#cltsReference'},
)
t.common_props['dc:description'] = \
"The ParameterTable lists IPA phones which appear in the DoReCo corpus (if a " \
"correspondence to the X-Sampa representation could be determined). If possible, IPA " \
"phones are linked to CLTS' BIPA representation, giving access to the CLTS feature " \
"system."
cldf[t, 'Name'].common_props['dc:description'] = "The IPA representation of the sound."
cldf[t, 'CLTS_ID'].common_props['dc:description'] = \
"CLTS ID of the sound, i.e. the underscore-separated ordered list of features of the " \
"sound."
cldf.add_table(
'speakers.csv',
{'name': 'ID', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#id'},
{
'name': 'Language_ID',
'datatype': 'string',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#languageReference'},
{
'name': 'age',
'dc:description': 'Speaker age. See also age_assignment_certain.',
'datatype': 'integer'},
{'name': 'age_assignment_certain',
'datatype': {'base': 'string', 'format': 'certain|approximate'},
},
{
'name': 'sex',
'dc:description': 'Speaker sex.',
'datatype': {'base': 'string', 'format': 'm|f'}},
)
t = cldf.add_table(
'phones.csv',
{'name': 'ph_ID', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#id'},
{
'name': 'ph',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#name',
'dc:description': 'See the description of the Token_Type column.',
'datatype': 'string'},
{
'name': 'IPA',
'dc:description': "Link to corresponding IPA phoneme, with details given in ParameterTable",
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#parameterReference'},
{
'name': 'u_ID',
'dc:description': 'Utterance ID. Utterances are words/phones delimited by silent pauses.',
'datatype': 'string'},
{
'name': 'Token_Type',
'dc:description': """\
Not all rows in this table correspond to actual phones. If a row does the Token_Type is 'xsampa'
and the `ph` column holds the X-SAMPA representation of the phone, otherwise it is a 'pause' or a
'label'.
Labels consist of two opening brackets, the label proper, a closing bracket, the content (optional),
and another closing bracket, e.g. `<<ui>word>`. Labels may also appear on their own if the content
is not known, e.g. `<<ui>>`. Valid proper labels are
- fp: Filled pause
- fs: False start
- pr: Prolongation
- fm: Foreign material
- sg: Singing
- bc: Backchannel
- id: Ideophone
- on: Onomatopoeic
- wip: Word-internal pause
- ui: Unidentifiable
Silent pauses are marked by a special symbol, `<p:>`. The location of silent pauses is manually
checked by the DoReCo team, while the symbol itself is inserted by the WebMAUS service. Unlike
labels, the <p:> symbol has only one of each bracket, and no other content may be included in it.\
""",
'datatype': {'base': 'string', 'format': 'label|pause|xsampa'}},
{
'name': 'start',
'dc:description': 'Start of the phone in the linked sound file in (floating point) seconds.',
'datatype': 'decimal'},
{
'name': 'end',
'dc:description': 'End of the phone in the linked sound file in (floating point) seconds.',
'datatype': 'decimal'},
{
'name': 'duration',
'dc:description': 'Duration of the phone in the linked sound file in (floating point) seconds.',
'datatype': 'decimal'},
{
'name': 'wd_ID',
'dc:description': 'Link to corresponding word.',
'datatype': 'string',
},
)
t.common_props['dc:description'] = 'This table lists individual, time-aligned phones.'
cldf.add_table(
'words.csv',
{'name': 'wd_ID', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#id'},
{
'name': 'wd',
'dc:description': 'The word form transcribed into orthography.',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#name',
'datatype': 'string',
},
{
'name': 'Language_ID',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#languageReference',
'datatype': 'string',
},
{
'name': 'File_ID',
'dc:description': 'Link to the audio file to which start and end markers pertain.',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#mediaReference',
'datatype': 'string',
},
{'name': 'Speaker_ID', 'datatype': 'string'},
{
'name': 'start',
'dc:description': 'Start of the word in the linked sound file in (floating point) seconds.',
'datatype': 'decimal',
},
{
'name': 'end',
'dc:description': 'End of the word in the linked sound file in (floating point) seconds.',
'datatype': 'decimal',
},
{
'name': 'duration',
'dc:description': 'Duration of the word in the linked sound file in (floating point) seconds.',
'datatype': 'decimal',
},
{
'name': 'ref',
'datatype': 'string',
},
{
'name': 'Example_ID',
'dc:description': 'Words that appear in glossed utterances are linked to an Example.',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#exampleReference',
},
{
'name': 'mb',
'separator': ' ',
'datatype': 'string',
},
{
'name': 'ps',
'separator': ' ',
'datatype': 'string',
},
{
'name': 'gl',
'separator': ' ',
'datatype': 'string',
},
)
t = cldf.add_table(
'glosses.csv',
{'name': 'ID', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#id'},
{'name': 'Gloss', 'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#name'},
{
'name': 'LGR',
'dc:description': 'Flag, signaling whether a gloss abbreviation is a standard, '
'Leipzig-Glossing-Rules abbreviation.',
'datatype': 'boolean',
},
{'name': 'Meaning', 'datatype': 'string'},
{
'name': 'Glottocode',
'propertyUrl': 'http://cldf.clld.org/v1.0/terms.rdf#languageReference',
'datatype': 'string',
}
)
t.common_props['dc:description'] = 'Gloss abbreviations used for glosses in a corpus.'
cldf.add_foreign_key('ContributionTable', 'ID', 'LanguageTable', 'ID')
cldf.add_foreign_key('phones.csv', 'wd_ID', 'words.csv', 'wd_ID')
cldf.add_foreign_key('words.csv', 'Speaker_ID', 'speakers.csv', 'ID')