|
1
|
|
|
# Copyright 2019-2020 by Christopher C. Little. |
|
2
|
|
|
# This file is part of Abydos. |
|
3
|
|
|
# |
|
4
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
|
5
|
|
|
# it under the terms of the GNU General Public License as published by |
|
6
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
|
7
|
|
|
# (at your option) any later version. |
|
8
|
|
|
# |
|
9
|
|
|
# Abydos is distributed in the hope that it will be useful, |
|
10
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12
|
|
|
# GNU General Public License for more details. |
|
13
|
|
|
# |
|
14
|
|
|
# You should have received a copy of the GNU General Public License |
|
15
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
|
16
|
|
|
|
|
17
|
|
|
"""abydos.distance._aline. |
|
18
|
|
|
|
|
19
|
1 |
|
ALINE alignment, similarity, and distance |
|
20
|
|
|
""" |
|
21
|
|
|
|
|
22
|
|
|
from copy import deepcopy |
|
23
|
|
|
from typing import Any, Callable, Dict, List, Tuple, Union, cast |
|
24
|
1 |
|
|
|
25
|
|
|
from numpy import float_, inf, zeros |
|
26
|
|
|
|
|
27
|
|
|
from ._distance import _Distance |
|
28
|
|
|
|
|
29
|
|
|
__all__ = ['ALINE'] |
|
30
|
|
|
|
|
31
|
1 |
|
|
|
32
|
|
|
class ALINE(_Distance): |
|
33
|
1 |
|
r"""ALINE alignment, similarity, and distance. |
|
34
|
1 |
|
|
|
35
|
1 |
|
ALINE alignment was developed by |
|
36
|
|
|
:cite:`Kondrak:2000,Kondrak:2002,Downey:2008`, and establishes an |
|
37
|
1 |
|
alignment algorithm based on multivalued phonetic features and feature |
|
38
|
|
|
salience weights. Along with the alignment itself, the algorithm produces a |
|
39
|
1 |
|
term similarity score. |
|
40
|
|
|
|
|
41
|
|
|
:cite:`Downey:2008` develops ALINE's similarity score into a similarity |
|
42
|
1 |
|
measure & distance measure: |
|
43
|
|
|
|
|
44
|
|
|
.. math:: |
|
45
|
|
|
|
|
46
|
|
|
sim_{ALINE} = \frac{2 \dot score_{ALINE}(src, tar)} |
|
47
|
|
|
{score_{ALINE}(src, src) + score_{ALINE}(tar, tar)} |
|
48
|
|
|
|
|
49
|
|
|
However, because the average of the two self-similarity scores is not |
|
50
|
|
|
guaranteed to be greater than or equal to the similarity score between |
|
51
|
|
|
the two strings, by default, this formula is not used here in order to |
|
52
|
|
|
guarantee that the similarity measure is bounded to [0, 1]. Instead, |
|
53
|
|
|
Kondrak's similarity measure is employed: |
|
54
|
|
|
|
|
55
|
|
|
.. math:: |
|
56
|
|
|
|
|
57
|
|
|
sim_{ALINE} = \frac{score_{ALINE}(src, tar)} |
|
58
|
|
|
{max(score_{ALINE}(src, src), score_{ALINE}(tar, tar))} |
|
59
|
|
|
|
|
60
|
|
|
|
|
61
|
|
|
.. versionadded:: 0.4.0 |
|
62
|
|
|
""" |
|
63
|
|
|
|
|
64
|
|
|
# The three dicts below are mostly copied from NLTK's implementation |
|
65
|
|
|
# https://www.nltk.org/_modules/nltk/metrics/aline.html |
|
66
|
|
|
# But values have been returned, as much as possible to the reference |
|
67
|
|
|
# values supplied in Kondrak's paper. |
|
68
|
|
|
feature_weights = { |
|
69
|
|
|
# place |
|
70
|
|
|
'bilabial': 1.0, |
|
71
|
|
|
'labiodental': 0.95, |
|
72
|
|
|
'dental': 0.9, |
|
73
|
|
|
'alveolar': 0.85, |
|
74
|
|
|
'retroflex': 0.8, |
|
75
|
|
|
'palato-alveolar': 0.75, |
|
76
|
|
|
'palatal': 0.7, |
|
77
|
|
|
'velar': 0.6, |
|
78
|
1 |
|
'uvular': 0.5, |
|
79
|
|
|
'pharyngeal': 0.3, |
|
80
|
|
|
'glottal': 0.1, |
|
81
|
|
|
# manner |
|
82
|
|
|
'stop': 1.0, |
|
83
|
|
|
'affricate': 0.9, |
|
84
|
|
|
'fricative': 0.8, |
|
85
|
|
|
'approximant': 0.6, |
|
86
|
|
|
'trill': 0.55, # not in original |
|
87
|
|
|
'tap': 0.5, # not in original |
|
88
|
|
|
'high vowel': 0.4, |
|
89
|
|
|
'mid vowel': 0.2, |
|
90
|
|
|
'low vowel': 0.0, |
|
91
|
|
|
# high |
|
92
|
|
|
'high': 1.0, |
|
93
|
|
|
'mid': 0.5, |
|
94
|
|
|
'low': 0.0, |
|
95
|
|
|
# back |
|
96
|
|
|
'front': 1.0, |
|
97
|
|
|
'central': 0.5, |
|
98
|
|
|
'back': 0.0, |
|
99
|
|
|
# binary features |
|
100
|
|
|
'plus': 1.0, |
|
101
|
|
|
'minus': 0.0, |
|
102
|
|
|
} |
|
103
|
|
|
|
|
104
|
|
|
v_features = { |
|
105
|
|
|
'syllabic', |
|
106
|
|
|
'nasal', |
|
107
|
|
|
'retroflex', |
|
108
|
|
|
'high', |
|
109
|
|
|
'back', |
|
110
|
|
|
'round', |
|
111
|
|
|
'long', |
|
112
|
|
|
} |
|
113
|
|
|
c_features = { |
|
114
|
1 |
|
'syllabic', |
|
115
|
|
|
'manner', |
|
116
|
|
|
'voice', |
|
117
|
|
|
'nasal', |
|
118
|
|
|
'retroflex', |
|
119
|
|
|
'lateral', |
|
120
|
|
|
'aspirated', |
|
121
|
|
|
'place', |
|
122
|
|
|
} |
|
123
|
1 |
|
|
|
124
|
|
|
salience = { |
|
125
|
|
|
'syllabic': 5, |
|
126
|
|
|
'voice': 10, |
|
127
|
|
|
'lateral': 10, |
|
128
|
|
|
'high': 5, |
|
129
|
|
|
'manner': 50, |
|
130
|
|
|
'long': 1, |
|
131
|
|
|
'place': 40, |
|
132
|
|
|
'nasal': 10, |
|
133
|
|
|
'aspirated': 5, |
|
134
|
1 |
|
'back': 5, |
|
135
|
|
|
'retroflex': 10, |
|
136
|
|
|
'round': 5, |
|
137
|
|
|
} |
|
138
|
|
|
|
|
139
|
|
|
phones_ipa = { |
|
140
|
|
|
'p': { |
|
141
|
|
|
'place': 'bilabial', |
|
142
|
|
|
'manner': 'stop', |
|
143
|
|
|
'syllabic': 'minus', |
|
144
|
|
|
'voice': 'minus', |
|
145
|
|
|
'nasal': 'minus', |
|
146
|
|
|
'retroflex': 'minus', |
|
147
|
|
|
'lateral': 'minus', |
|
148
|
|
|
'aspirated': 'minus', |
|
149
|
1 |
|
}, |
|
150
|
|
|
'b': { |
|
151
|
|
|
'place': 'bilabial', |
|
152
|
|
|
'manner': 'stop', |
|
153
|
|
|
'syllabic': 'minus', |
|
154
|
|
|
'voice': 'plus', |
|
155
|
|
|
'nasal': 'minus', |
|
156
|
|
|
'retroflex': 'minus', |
|
157
|
|
|
'lateral': 'minus', |
|
158
|
|
|
'aspirated': 'minus', |
|
159
|
|
|
}, |
|
160
|
|
|
't': { |
|
161
|
|
|
'place': 'alveolar', |
|
162
|
|
|
'manner': 'stop', |
|
163
|
|
|
'syllabic': 'minus', |
|
164
|
|
|
'voice': 'minus', |
|
165
|
|
|
'nasal': 'minus', |
|
166
|
|
|
'retroflex': 'minus', |
|
167
|
|
|
'lateral': 'minus', |
|
168
|
|
|
'aspirated': 'minus', |
|
169
|
|
|
}, |
|
170
|
|
|
'd': { |
|
171
|
|
|
'place': 'alveolar', |
|
172
|
|
|
'manner': 'stop', |
|
173
|
|
|
'syllabic': 'minus', |
|
174
|
|
|
'voice': 'plus', |
|
175
|
|
|
'nasal': 'minus', |
|
176
|
|
|
'retroflex': 'minus', |
|
177
|
|
|
'lateral': 'minus', |
|
178
|
|
|
'aspirated': 'minus', |
|
179
|
|
|
}, |
|
180
|
|
|
'ʈ': { |
|
181
|
|
|
'place': 'retroflex', |
|
182
|
|
|
'manner': 'stop', |
|
183
|
|
|
'syllabic': 'minus', |
|
184
|
|
|
'voice': 'minus', |
|
185
|
|
|
'nasal': 'minus', |
|
186
|
|
|
'retroflex': 'plus', |
|
187
|
|
|
'lateral': 'minus', |
|
188
|
|
|
'aspirated': 'minus', |
|
189
|
|
|
}, |
|
190
|
|
|
'ɖ': { |
|
191
|
|
|
'place': 'retroflex', |
|
192
|
|
|
'manner': 'stop', |
|
193
|
|
|
'syllabic': 'minus', |
|
194
|
|
|
'voice': 'plus', |
|
195
|
|
|
'nasal': 'minus', |
|
196
|
|
|
'retroflex': 'plus', |
|
197
|
|
|
'lateral': 'minus', |
|
198
|
|
|
'aspirated': 'minus', |
|
199
|
|
|
}, |
|
200
|
|
|
'c': { |
|
201
|
|
|
'place': 'palatal', |
|
202
|
|
|
'manner': 'stop', |
|
203
|
|
|
'syllabic': 'minus', |
|
204
|
|
|
'voice': 'minus', |
|
205
|
|
|
'nasal': 'minus', |
|
206
|
|
|
'retroflex': 'minus', |
|
207
|
|
|
'lateral': 'minus', |
|
208
|
|
|
'aspirated': 'minus', |
|
209
|
|
|
}, |
|
210
|
|
|
'ɟ': { |
|
211
|
|
|
'place': 'palatal', |
|
212
|
|
|
'manner': 'stop', |
|
213
|
|
|
'syllabic': 'minus', |
|
214
|
|
|
'voice': 'plus', |
|
215
|
|
|
'nasal': 'minus', |
|
216
|
|
|
'retroflex': 'minus', |
|
217
|
|
|
'lateral': 'minus', |
|
218
|
|
|
'aspirated': 'minus', |
|
219
|
|
|
}, |
|
220
|
|
|
'k': { |
|
221
|
|
|
'place': 'velar', |
|
222
|
|
|
'manner': 'stop', |
|
223
|
|
|
'syllabic': 'minus', |
|
224
|
|
|
'voice': 'minus', |
|
225
|
|
|
'nasal': 'minus', |
|
226
|
|
|
'retroflex': 'minus', |
|
227
|
|
|
'lateral': 'minus', |
|
228
|
|
|
'aspirated': 'minus', |
|
229
|
|
|
}, |
|
230
|
|
|
'g': { |
|
231
|
|
|
'place': 'velar', |
|
232
|
|
|
'manner': 'stop', |
|
233
|
|
|
'syllabic': 'minus', |
|
234
|
|
|
'voice': 'plus', |
|
235
|
|
|
'nasal': 'minus', |
|
236
|
|
|
'retroflex': 'minus', |
|
237
|
|
|
'lateral': 'minus', |
|
238
|
|
|
'aspirated': 'minus', |
|
239
|
|
|
}, |
|
240
|
|
|
'q': { |
|
241
|
|
|
'place': 'uvular', |
|
242
|
|
|
'manner': 'stop', |
|
243
|
|
|
'syllabic': 'minus', |
|
244
|
|
|
'voice': 'minus', |
|
245
|
|
|
'nasal': 'minus', |
|
246
|
|
|
'retroflex': 'minus', |
|
247
|
|
|
'lateral': 'minus', |
|
248
|
|
|
'aspirated': 'minus', |
|
249
|
|
|
}, |
|
250
|
|
|
'ɢ': { |
|
251
|
|
|
'place': 'uvular', |
|
252
|
|
|
'manner': 'stop', |
|
253
|
|
|
'syllabic': 'minus', |
|
254
|
|
|
'voice': 'plus', |
|
255
|
|
|
'nasal': 'minus', |
|
256
|
|
|
'retroflex': 'minus', |
|
257
|
|
|
'lateral': 'minus', |
|
258
|
|
|
'aspirated': 'minus', |
|
259
|
|
|
}, |
|
260
|
|
|
'ʔ': { |
|
261
|
|
|
'place': 'glottal', |
|
262
|
|
|
'manner': 'stop', |
|
263
|
|
|
'syllabic': 'minus', |
|
264
|
|
|
'voice': 'minus', |
|
265
|
|
|
'nasal': 'minus', |
|
266
|
|
|
'retroflex': 'minus', |
|
267
|
|
|
'lateral': 'minus', |
|
268
|
|
|
'aspirated': 'minus', |
|
269
|
|
|
}, |
|
270
|
|
|
'm': { |
|
271
|
|
|
'place': 'bilabial', |
|
272
|
|
|
'manner': 'stop', |
|
273
|
|
|
'syllabic': 'minus', |
|
274
|
|
|
'voice': 'plus', |
|
275
|
|
|
'nasal': 'plus', |
|
276
|
|
|
'retroflex': 'minus', |
|
277
|
|
|
'lateral': 'minus', |
|
278
|
|
|
'aspirated': 'minus', |
|
279
|
|
|
}, |
|
280
|
|
|
'ɱ': { |
|
281
|
|
|
'place': 'labiodental', |
|
282
|
|
|
'manner': 'stop', |
|
283
|
|
|
'syllabic': 'minus', |
|
284
|
|
|
'voice': 'plus', |
|
285
|
|
|
'nasal': 'plus', |
|
286
|
|
|
'retroflex': 'minus', |
|
287
|
|
|
'lateral': 'minus', |
|
288
|
|
|
'aspirated': 'minus', |
|
289
|
|
|
}, |
|
290
|
|
|
'n': { |
|
291
|
|
|
'place': 'alveolar', |
|
292
|
|
|
'manner': 'stop', |
|
293
|
|
|
'syllabic': 'minus', |
|
294
|
|
|
'voice': 'plus', |
|
295
|
|
|
'nasal': 'plus', |
|
296
|
|
|
'retroflex': 'minus', |
|
297
|
|
|
'lateral': 'minus', |
|
298
|
|
|
'aspirated': 'minus', |
|
299
|
|
|
}, |
|
300
|
|
|
'ɳ': { |
|
301
|
|
|
'place': 'retroflex', |
|
302
|
|
|
'manner': 'stop', |
|
303
|
|
|
'syllabic': 'minus', |
|
304
|
|
|
'voice': 'plus', |
|
305
|
|
|
'nasal': 'plus', |
|
306
|
|
|
'retroflex': 'plus', |
|
307
|
|
|
'lateral': 'minus', |
|
308
|
|
|
'aspirated': 'minus', |
|
309
|
|
|
}, |
|
310
|
|
|
'ɲ': { |
|
311
|
|
|
'place': 'palatal', |
|
312
|
|
|
'manner': 'stop', |
|
313
|
|
|
'syllabic': 'minus', |
|
314
|
|
|
'voice': 'plus', |
|
315
|
|
|
'nasal': 'plus', |
|
316
|
|
|
'retroflex': 'minus', |
|
317
|
|
|
'lateral': 'minus', |
|
318
|
|
|
'aspirated': 'minus', |
|
319
|
|
|
}, |
|
320
|
|
|
'ŋ': { |
|
321
|
|
|
'place': 'velar', |
|
322
|
|
|
'manner': 'stop', |
|
323
|
|
|
'syllabic': 'minus', |
|
324
|
|
|
'voice': 'plus', |
|
325
|
|
|
'nasal': 'plus', |
|
326
|
|
|
'retroflex': 'minus', |
|
327
|
|
|
'lateral': 'minus', |
|
328
|
|
|
'aspirated': 'minus', |
|
329
|
|
|
}, |
|
330
|
|
|
'ɴ': { |
|
331
|
|
|
'place': 'uvular', |
|
332
|
|
|
'manner': 'stop', |
|
333
|
|
|
'syllabic': 'minus', |
|
334
|
|
|
'voice': 'plus', |
|
335
|
|
|
'nasal': 'plus', |
|
336
|
|
|
'retroflex': 'minus', |
|
337
|
|
|
'lateral': 'minus', |
|
338
|
|
|
'aspirated': 'minus', |
|
339
|
|
|
}, |
|
340
|
|
|
'ʙ': { |
|
341
|
|
|
'place': 'bilabial', |
|
342
|
|
|
'manner': 'trill', |
|
343
|
|
|
'syllabic': 'minus', |
|
344
|
|
|
'voice': 'plus', |
|
345
|
|
|
'nasal': 'minus', |
|
346
|
|
|
'retroflex': 'minus', |
|
347
|
|
|
'lateral': 'minus', |
|
348
|
|
|
'aspirated': 'minus', |
|
349
|
|
|
}, |
|
350
|
|
|
'r': { |
|
351
|
|
|
'place': 'alveolar', |
|
352
|
|
|
'manner': 'trill', |
|
353
|
|
|
'syllabic': 'minus', |
|
354
|
|
|
'voice': 'plus', |
|
355
|
|
|
'nasal': 'minus', |
|
356
|
|
|
'retroflex': 'plus', |
|
357
|
|
|
'lateral': 'minus', |
|
358
|
|
|
'aspirated': 'minus', |
|
359
|
|
|
}, |
|
360
|
|
|
'ʀ': { |
|
361
|
|
|
'place': 'uvular', |
|
362
|
|
|
'manner': 'trill', |
|
363
|
|
|
'syllabic': 'minus', |
|
364
|
|
|
'voice': 'plus', |
|
365
|
|
|
'nasal': 'minus', |
|
366
|
|
|
'retroflex': 'minus', |
|
367
|
|
|
'lateral': 'minus', |
|
368
|
|
|
'aspirated': 'minus', |
|
369
|
|
|
}, |
|
370
|
|
|
'ɾ': { |
|
371
|
|
|
'place': 'alveolar', |
|
372
|
|
|
'manner': 'tap', |
|
373
|
|
|
'syllabic': 'minus', |
|
374
|
|
|
'voice': 'plus', |
|
375
|
|
|
'nasal': 'minus', |
|
376
|
|
|
'retroflex': 'minus', |
|
377
|
|
|
'lateral': 'minus', |
|
378
|
|
|
'aspirated': 'minus', |
|
379
|
|
|
}, |
|
380
|
|
|
'ɽ': { |
|
381
|
|
|
'place': 'retroflex', |
|
382
|
|
|
'manner': 'tap', |
|
383
|
|
|
'syllabic': 'minus', |
|
384
|
|
|
'voice': 'plus', |
|
385
|
|
|
'nasal': 'minus', |
|
386
|
|
|
'retroflex': 'plus', |
|
387
|
|
|
'lateral': 'minus', |
|
388
|
|
|
'aspirated': 'minus', |
|
389
|
|
|
}, |
|
390
|
|
|
'ɸ': { |
|
391
|
|
|
'place': 'bilabial', |
|
392
|
|
|
'manner': 'fricative', |
|
393
|
|
|
'syllabic': 'minus', |
|
394
|
|
|
'voice': 'minus', |
|
395
|
|
|
'nasal': 'minus', |
|
396
|
|
|
'retroflex': 'minus', |
|
397
|
|
|
'lateral': 'minus', |
|
398
|
|
|
'aspirated': 'minus', |
|
399
|
|
|
}, |
|
400
|
|
|
'β': { |
|
401
|
|
|
'place': 'bilabial', |
|
402
|
|
|
'manner': 'fricative', |
|
403
|
|
|
'syllabic': 'minus', |
|
404
|
|
|
'voice': 'plus', |
|
405
|
|
|
'nasal': 'minus', |
|
406
|
|
|
'retroflex': 'minus', |
|
407
|
|
|
'lateral': 'minus', |
|
408
|
|
|
'aspirated': 'minus', |
|
409
|
|
|
}, |
|
410
|
|
|
'f': { |
|
411
|
|
|
'place': 'labiodental', |
|
412
|
|
|
'manner': 'fricative', |
|
413
|
|
|
'syllabic': 'minus', |
|
414
|
|
|
'voice': 'minus', |
|
415
|
|
|
'nasal': 'minus', |
|
416
|
|
|
'retroflex': 'minus', |
|
417
|
|
|
'lateral': 'minus', |
|
418
|
|
|
'aspirated': 'minus', |
|
419
|
|
|
}, |
|
420
|
|
|
'v': { |
|
421
|
|
|
'place': 'labiodental', |
|
422
|
|
|
'manner': 'fricative', |
|
423
|
|
|
'syllabic': 'minus', |
|
424
|
|
|
'voice': 'plus', |
|
425
|
|
|
'nasal': 'minus', |
|
426
|
|
|
'retroflex': 'minus', |
|
427
|
|
|
'lateral': 'minus', |
|
428
|
|
|
'aspirated': 'minus', |
|
429
|
|
|
}, |
|
430
|
|
|
'θ': { |
|
431
|
|
|
'place': 'dental', |
|
432
|
|
|
'manner': 'fricative', |
|
433
|
|
|
'syllabic': 'minus', |
|
434
|
|
|
'voice': 'minus', |
|
435
|
|
|
'nasal': 'minus', |
|
436
|
|
|
'retroflex': 'minus', |
|
437
|
|
|
'lateral': 'minus', |
|
438
|
|
|
'aspirated': 'minus', |
|
439
|
|
|
}, |
|
440
|
|
|
'ð': { |
|
441
|
|
|
'place': 'dental', |
|
442
|
|
|
'manner': 'fricative', |
|
443
|
|
|
'syllabic': 'minus', |
|
444
|
|
|
'voice': 'plus', |
|
445
|
|
|
'nasal': 'minus', |
|
446
|
|
|
'retroflex': 'minus', |
|
447
|
|
|
'lateral': 'minus', |
|
448
|
|
|
'aspirated': 'minus', |
|
449
|
|
|
}, |
|
450
|
|
|
's': { |
|
451
|
|
|
'place': 'alveolar', |
|
452
|
|
|
'manner': 'fricative', |
|
453
|
|
|
'syllabic': 'minus', |
|
454
|
|
|
'voice': 'minus', |
|
455
|
|
|
'nasal': 'minus', |
|
456
|
|
|
'retroflex': 'minus', |
|
457
|
|
|
'lateral': 'minus', |
|
458
|
|
|
'aspirated': 'minus', |
|
459
|
|
|
}, |
|
460
|
|
|
'z': { |
|
461
|
|
|
'place': 'alveolar', |
|
462
|
|
|
'manner': 'fricative', |
|
463
|
|
|
'syllabic': 'minus', |
|
464
|
|
|
'voice': 'plus', |
|
465
|
|
|
'nasal': 'minus', |
|
466
|
|
|
'retroflex': 'minus', |
|
467
|
|
|
'lateral': 'minus', |
|
468
|
|
|
'aspirated': 'minus', |
|
469
|
|
|
}, |
|
470
|
|
|
'ʃ': { |
|
471
|
|
|
'place': 'palato-alveolar', |
|
472
|
|
|
'manner': 'fricative', |
|
473
|
|
|
'syllabic': 'minus', |
|
474
|
|
|
'voice': 'minus', |
|
475
|
|
|
'nasal': 'minus', |
|
476
|
|
|
'retroflex': 'minus', |
|
477
|
|
|
'lateral': 'minus', |
|
478
|
|
|
'aspirated': 'minus', |
|
479
|
|
|
}, |
|
480
|
|
|
'ʒ': { |
|
481
|
|
|
'place': 'palato-alveolar', |
|
482
|
|
|
'manner': 'fricative', |
|
483
|
|
|
'syllabic': 'minus', |
|
484
|
|
|
'voice': 'plus', |
|
485
|
|
|
'nasal': 'minus', |
|
486
|
|
|
'retroflex': 'minus', |
|
487
|
|
|
'lateral': 'minus', |
|
488
|
|
|
'aspirated': 'minus', |
|
489
|
|
|
}, |
|
490
|
|
|
'ʂ': { |
|
491
|
|
|
'place': 'retroflex', |
|
492
|
|
|
'manner': 'fricative', |
|
493
|
|
|
'syllabic': 'minus', |
|
494
|
|
|
'voice': 'minus', |
|
495
|
|
|
'nasal': 'minus', |
|
496
|
|
|
'retroflex': 'plus', |
|
497
|
|
|
'lateral': 'minus', |
|
498
|
|
|
'aspirated': 'minus', |
|
499
|
|
|
}, |
|
500
|
|
|
'ʐ': { |
|
501
|
|
|
'place': 'retroflex', |
|
502
|
|
|
'manner': 'fricative', |
|
503
|
|
|
'syllabic': 'minus', |
|
504
|
|
|
'voice': 'plus', |
|
505
|
|
|
'nasal': 'minus', |
|
506
|
|
|
'retroflex': 'plus', |
|
507
|
|
|
'lateral': 'minus', |
|
508
|
|
|
'aspirated': 'minus', |
|
509
|
|
|
}, |
|
510
|
|
|
'ç': { |
|
511
|
|
|
'place': 'palatal', |
|
512
|
|
|
'manner': 'fricative', |
|
513
|
|
|
'syllabic': 'minus', |
|
514
|
|
|
'voice': 'minus', |
|
515
|
|
|
'nasal': 'minus', |
|
516
|
|
|
'retroflex': 'minus', |
|
517
|
|
|
'lateral': 'minus', |
|
518
|
|
|
'aspirated': 'minus', |
|
519
|
|
|
}, |
|
520
|
|
|
'ʝ': { |
|
521
|
|
|
'place': 'palatal', |
|
522
|
|
|
'manner': 'fricative', |
|
523
|
|
|
'syllabic': 'minus', |
|
524
|
|
|
'voice': 'plus', |
|
525
|
|
|
'nasal': 'minus', |
|
526
|
|
|
'retroflex': 'minus', |
|
527
|
|
|
'lateral': 'minus', |
|
528
|
|
|
'aspirated': 'minus', |
|
529
|
|
|
}, |
|
530
|
|
|
'x': { |
|
531
|
|
|
'place': 'velar', |
|
532
|
|
|
'manner': 'fricative', |
|
533
|
|
|
'syllabic': 'minus', |
|
534
|
|
|
'voice': 'minus', |
|
535
|
|
|
'nasal': 'minus', |
|
536
|
|
|
'retroflex': 'minus', |
|
537
|
|
|
'lateral': 'minus', |
|
538
|
|
|
'aspirated': 'minus', |
|
539
|
|
|
}, |
|
540
|
|
|
'ɣ': { |
|
541
|
|
|
'place': 'velar', |
|
542
|
|
|
'manner': 'fricative', |
|
543
|
|
|
'syllabic': 'minus', |
|
544
|
|
|
'voice': 'plus', |
|
545
|
|
|
'nasal': 'minus', |
|
546
|
|
|
'retroflex': 'minus', |
|
547
|
|
|
'lateral': 'minus', |
|
548
|
|
|
'aspirated': 'minus', |
|
549
|
|
|
}, |
|
550
|
|
|
'χ': { |
|
551
|
|
|
'place': 'uvular', |
|
552
|
|
|
'manner': 'fricative', |
|
553
|
|
|
'syllabic': 'minus', |
|
554
|
|
|
'voice': 'minus', |
|
555
|
|
|
'nasal': 'minus', |
|
556
|
|
|
'retroflex': 'minus', |
|
557
|
|
|
'lateral': 'minus', |
|
558
|
|
|
'aspirated': 'minus', |
|
559
|
|
|
}, |
|
560
|
|
|
'ʁ': { |
|
561
|
|
|
'place': 'uvular', |
|
562
|
|
|
'manner': 'fricative', |
|
563
|
|
|
'syllabic': 'minus', |
|
564
|
|
|
'voice': 'plus', |
|
565
|
|
|
'nasal': 'minus', |
|
566
|
|
|
'retroflex': 'minus', |
|
567
|
|
|
'lateral': 'minus', |
|
568
|
|
|
'aspirated': 'minus', |
|
569
|
|
|
}, |
|
570
|
|
|
'ħ': { |
|
571
|
|
|
'place': 'pharyngeal', |
|
572
|
|
|
'manner': 'fricative', |
|
573
|
|
|
'syllabic': 'minus', |
|
574
|
|
|
'voice': 'minus', |
|
575
|
|
|
'nasal': 'minus', |
|
576
|
|
|
'retroflex': 'minus', |
|
577
|
|
|
'lateral': 'minus', |
|
578
|
|
|
'aspirated': 'minus', |
|
579
|
|
|
}, |
|
580
|
|
|
'ʕ': { |
|
581
|
|
|
'place': 'pharyngeal', |
|
582
|
|
|
'manner': 'fricative', |
|
583
|
|
|
'syllabic': 'minus', |
|
584
|
|
|
'voice': 'plus', |
|
585
|
|
|
'nasal': 'minus', |
|
586
|
|
|
'retroflex': 'minus', |
|
587
|
|
|
'lateral': 'minus', |
|
588
|
|
|
'aspirated': 'minus', |
|
589
|
|
|
}, |
|
590
|
|
|
'h': { |
|
591
|
|
|
'place': 'glottal', |
|
592
|
|
|
'manner': 'fricative', |
|
593
|
|
|
'syllabic': 'minus', |
|
594
|
|
|
'voice': 'minus', |
|
595
|
|
|
'nasal': 'minus', |
|
596
|
|
|
'retroflex': 'minus', |
|
597
|
|
|
'lateral': 'minus', |
|
598
|
|
|
'aspirated': 'minus', |
|
599
|
|
|
}, |
|
600
|
|
|
'ɦ': { |
|
601
|
|
|
'place': 'glottal', |
|
602
|
|
|
'manner': 'fricative', |
|
603
|
|
|
'syllabic': 'minus', |
|
604
|
|
|
'voice': 'plus', |
|
605
|
|
|
'nasal': 'minus', |
|
606
|
|
|
'retroflex': 'minus', |
|
607
|
|
|
'lateral': 'minus', |
|
608
|
|
|
'aspirated': 'minus', |
|
609
|
|
|
}, |
|
610
|
|
|
'ɬ': { |
|
611
|
|
|
'place': 'alveolar', |
|
612
|
|
|
'manner': 'fricative', |
|
613
|
|
|
'syllabic': 'minus', |
|
614
|
|
|
'voice': 'minus', |
|
615
|
|
|
'nasal': 'minus', |
|
616
|
|
|
'retroflex': 'minus', |
|
617
|
|
|
'lateral': 'plus', |
|
618
|
|
|
'aspirated': 'minus', |
|
619
|
|
|
}, |
|
620
|
|
|
'ɮ': { |
|
621
|
|
|
'place': 'alveolar', |
|
622
|
|
|
'manner': 'fricative', |
|
623
|
|
|
'syllabic': 'minus', |
|
624
|
|
|
'voice': 'plus', |
|
625
|
|
|
'nasal': 'minus', |
|
626
|
|
|
'retroflex': 'minus', |
|
627
|
|
|
'lateral': 'plus', |
|
628
|
|
|
'aspirated': 'minus', |
|
629
|
|
|
}, |
|
630
|
|
|
'ʋ': { |
|
631
|
|
|
'place': 'labiodental', |
|
632
|
|
|
'manner': 'approximant', |
|
633
|
|
|
'syllabic': 'minus', |
|
634
|
|
|
'voice': 'plus', |
|
635
|
|
|
'nasal': 'minus', |
|
636
|
|
|
'retroflex': 'minus', |
|
637
|
|
|
'lateral': 'minus', |
|
638
|
|
|
'aspirated': 'minus', |
|
639
|
|
|
}, |
|
640
|
|
|
'ɹ': { |
|
641
|
|
|
'place': 'alveolar', |
|
642
|
|
|
'manner': 'approximant', |
|
643
|
|
|
'syllabic': 'minus', |
|
644
|
|
|
'voice': 'plus', |
|
645
|
|
|
'nasal': 'minus', |
|
646
|
|
|
'retroflex': 'minus', |
|
647
|
|
|
'lateral': 'minus', |
|
648
|
|
|
'aspirated': 'minus', |
|
649
|
|
|
}, |
|
650
|
|
|
'ɻ': { |
|
651
|
|
|
'place': 'retroflex', |
|
652
|
|
|
'manner': 'approximant', |
|
653
|
|
|
'syllabic': 'minus', |
|
654
|
|
|
'voice': 'plus', |
|
655
|
|
|
'nasal': 'minus', |
|
656
|
|
|
'retroflex': 'plus', |
|
657
|
|
|
'lateral': 'minus', |
|
658
|
|
|
'aspirated': 'minus', |
|
659
|
|
|
}, |
|
660
|
|
|
'j': { |
|
661
|
|
|
'place': 'palatal', |
|
662
|
|
|
'manner': 'approximant', |
|
663
|
|
|
'syllabic': 'minus', |
|
664
|
|
|
'voice': 'plus', |
|
665
|
|
|
'nasal': 'minus', |
|
666
|
|
|
'retroflex': 'minus', |
|
667
|
|
|
'lateral': 'minus', |
|
668
|
|
|
'aspirated': 'minus', |
|
669
|
|
|
}, |
|
670
|
|
|
'ɰ': { |
|
671
|
|
|
'place': 'velar', |
|
672
|
|
|
'manner': 'approximant', |
|
673
|
|
|
'syllabic': 'minus', |
|
674
|
|
|
'voice': 'plus', |
|
675
|
|
|
'nasal': 'minus', |
|
676
|
|
|
'retroflex': 'minus', |
|
677
|
|
|
'lateral': 'minus', |
|
678
|
|
|
'aspirated': 'minus', |
|
679
|
|
|
}, |
|
680
|
|
|
'l': { |
|
681
|
|
|
'place': 'alveolar', |
|
682
|
|
|
'manner': 'approximant', |
|
683
|
|
|
'syllabic': 'minus', |
|
684
|
|
|
'voice': 'plus', |
|
685
|
|
|
'nasal': 'minus', |
|
686
|
|
|
'retroflex': 'minus', |
|
687
|
|
|
'lateral': 'plus', |
|
688
|
|
|
'aspirated': 'minus', |
|
689
|
|
|
}, |
|
690
|
|
|
'w': { |
|
691
|
|
|
'place': 'velar', |
|
692
|
|
|
'manner': 'approximant', |
|
693
|
|
|
'syllabic': 'minus', |
|
694
|
|
|
'voice': 'plus', |
|
695
|
|
|
'nasal': 'minus', |
|
696
|
|
|
'retroflex': 'minus', |
|
697
|
|
|
'lateral': 'minus', |
|
698
|
|
|
'aspirated': 'minus', |
|
699
|
|
|
'double': 'bilabial', |
|
700
|
|
|
}, |
|
701
|
|
|
'i': { |
|
702
|
|
|
'manner': 'high vowel', |
|
703
|
|
|
'syllabic': 'plus', |
|
704
|
|
|
'voice': 'plus', |
|
705
|
|
|
'nasal': 'minus', |
|
706
|
|
|
'retroflex': 'minus', |
|
707
|
|
|
'lateral': 'minus', |
|
708
|
|
|
'high': 'high', |
|
709
|
|
|
'back': 'front', |
|
710
|
|
|
'round': 'minus', |
|
711
|
|
|
'long': 'minus', |
|
712
|
|
|
'aspirated': 'minus', |
|
713
|
|
|
}, |
|
714
|
|
|
'y': { |
|
715
|
|
|
'manner': 'high vowel', |
|
716
|
|
|
'syllabic': 'plus', |
|
717
|
|
|
'voice': 'plus', |
|
718
|
|
|
'nasal': 'minus', |
|
719
|
|
|
'retroflex': 'minus', |
|
720
|
|
|
'lateral': 'minus', |
|
721
|
|
|
'high': 'high', |
|
722
|
|
|
'back': 'front', |
|
723
|
|
|
'round': 'plus', |
|
724
|
|
|
'long': 'minus', |
|
725
|
|
|
'aspirated': 'minus', |
|
726
|
|
|
}, |
|
727
|
|
|
'e': { |
|
728
|
|
|
'manner': 'mid vowel', |
|
729
|
|
|
'syllabic': 'plus', |
|
730
|
|
|
'voice': 'plus', |
|
731
|
|
|
'nasal': 'minus', |
|
732
|
|
|
'retroflex': 'minus', |
|
733
|
|
|
'lateral': 'minus', |
|
734
|
|
|
'high': 'mid', |
|
735
|
|
|
'back': 'front', |
|
736
|
|
|
'round': 'minus', |
|
737
|
|
|
'long': 'minus', |
|
738
|
|
|
'aspirated': 'minus', |
|
739
|
|
|
}, |
|
740
|
|
|
'ø': { |
|
741
|
|
|
'manner': 'mid vowel', |
|
742
|
|
|
'syllabic': 'plus', |
|
743
|
|
|
'voice': 'plus', |
|
744
|
|
|
'nasal': 'minus', |
|
745
|
|
|
'retroflex': 'minus', |
|
746
|
|
|
'lateral': 'minus', |
|
747
|
|
|
'high': 'mid', |
|
748
|
|
|
'back': 'front', |
|
749
|
|
|
'round': 'plus', |
|
750
|
|
|
'long': 'minus', |
|
751
|
|
|
'aspirated': 'minus', |
|
752
|
|
|
}, |
|
753
|
|
|
'ɛ': { |
|
754
|
|
|
'manner': 'mid vowel', |
|
755
|
|
|
'syllabic': 'plus', |
|
756
|
|
|
'voice': 'plus', |
|
757
|
|
|
'nasal': 'minus', |
|
758
|
|
|
'retroflex': 'minus', |
|
759
|
|
|
'lateral': 'minus', |
|
760
|
|
|
'high': 'mid', |
|
761
|
|
|
'back': 'front', |
|
762
|
|
|
'round': 'minus', |
|
763
|
|
|
'long': 'minus', |
|
764
|
|
|
'aspirated': 'minus', |
|
765
|
|
|
}, |
|
766
|
|
|
'œ': { |
|
767
|
|
|
'manner': 'mid vowel', |
|
768
|
|
|
'syllabic': 'plus', |
|
769
|
|
|
'voice': 'plus', |
|
770
|
|
|
'nasal': 'minus', |
|
771
|
|
|
'retroflex': 'minus', |
|
772
|
|
|
'lateral': 'minus', |
|
773
|
|
|
'high': 'mid', |
|
774
|
|
|
'back': 'front', |
|
775
|
|
|
'round': 'plus', |
|
776
|
|
|
'long': 'minus', |
|
777
|
|
|
'aspirated': 'minus', |
|
778
|
|
|
}, |
|
779
|
|
|
'æ': { |
|
780
|
|
|
'manner': 'low vowel', |
|
781
|
|
|
'syllabic': 'plus', |
|
782
|
|
|
'voice': 'plus', |
|
783
|
|
|
'nasal': 'minus', |
|
784
|
|
|
'retroflex': 'minus', |
|
785
|
|
|
'lateral': 'minus', |
|
786
|
|
|
'high': 'low', |
|
787
|
|
|
'back': 'front', |
|
788
|
|
|
'round': 'minus', |
|
789
|
|
|
'long': 'minus', |
|
790
|
|
|
'aspirated': 'minus', |
|
791
|
|
|
}, |
|
792
|
|
|
'a': { |
|
793
|
|
|
'manner': 'low vowel', |
|
794
|
|
|
'syllabic': 'plus', |
|
795
|
|
|
'voice': 'plus', |
|
796
|
|
|
'nasal': 'minus', |
|
797
|
|
|
'retroflex': 'minus', |
|
798
|
|
|
'lateral': 'minus', |
|
799
|
|
|
'high': 'low', |
|
800
|
|
|
'back': 'front', |
|
801
|
|
|
'round': 'minus', |
|
802
|
|
|
'long': 'minus', |
|
803
|
|
|
'aspirated': 'minus', |
|
804
|
|
|
}, |
|
805
|
|
|
'ɨ': { |
|
806
|
|
|
'manner': 'high vowel', |
|
807
|
|
|
'syllabic': 'plus', |
|
808
|
|
|
'voice': 'plus', |
|
809
|
|
|
'nasal': 'minus', |
|
810
|
|
|
'retroflex': 'minus', |
|
811
|
|
|
'lateral': 'minus', |
|
812
|
|
|
'high': 'high', |
|
813
|
|
|
'back': 'central', |
|
814
|
|
|
'round': 'minus', |
|
815
|
|
|
'long': 'minus', |
|
816
|
|
|
'aspirated': 'minus', |
|
817
|
|
|
}, |
|
818
|
|
|
'ʉ': { |
|
819
|
|
|
'manner': 'high vowel', |
|
820
|
|
|
'syllabic': 'plus', |
|
821
|
|
|
'voice': 'plus', |
|
822
|
|
|
'nasal': 'minus', |
|
823
|
|
|
'retroflex': 'minus', |
|
824
|
|
|
'lateral': 'minus', |
|
825
|
|
|
'high': 'high', |
|
826
|
|
|
'back': 'central', |
|
827
|
|
|
'round': 'plus', |
|
828
|
|
|
'long': 'minus', |
|
829
|
|
|
'aspirated': 'minus', |
|
830
|
|
|
}, |
|
831
|
|
|
'ə': { |
|
832
|
|
|
'manner': 'mid vowel', |
|
833
|
|
|
'syllabic': 'plus', |
|
834
|
|
|
'voice': 'plus', |
|
835
|
|
|
'nasal': 'minus', |
|
836
|
|
|
'retroflex': 'minus', |
|
837
|
|
|
'lateral': 'minus', |
|
838
|
|
|
'high': 'mid', |
|
839
|
|
|
'back': 'central', |
|
840
|
|
|
'round': 'minus', |
|
841
|
|
|
'long': 'minus', |
|
842
|
|
|
'aspirated': 'minus', |
|
843
|
|
|
}, |
|
844
|
|
|
'u': { |
|
845
|
|
|
'manner': 'high vowel', |
|
846
|
|
|
'syllabic': 'plus', |
|
847
|
|
|
'voice': 'plus', |
|
848
|
|
|
'nasal': 'minus', |
|
849
|
|
|
'retroflex': 'minus', |
|
850
|
|
|
'lateral': 'minus', |
|
851
|
|
|
'high': 'high', |
|
852
|
|
|
'back': 'back', |
|
853
|
|
|
'round': 'plus', |
|
854
|
|
|
'long': 'minus', |
|
855
|
|
|
'aspirated': 'minus', |
|
856
|
|
|
}, |
|
857
|
|
|
'o': { |
|
858
|
|
|
'manner': 'mid vowel', |
|
859
|
|
|
'syllabic': 'plus', |
|
860
|
|
|
'voice': 'plus', |
|
861
|
|
|
'nasal': 'minus', |
|
862
|
|
|
'retroflex': 'minus', |
|
863
|
|
|
'lateral': 'minus', |
|
864
|
|
|
'high': 'mid', |
|
865
|
|
|
'back': 'back', |
|
866
|
|
|
'round': 'plus', |
|
867
|
|
|
'long': 'minus', |
|
868
|
|
|
'aspirated': 'minus', |
|
869
|
|
|
}, |
|
870
|
|
|
'ɔ': { |
|
871
|
|
|
'manner': 'mid vowel', |
|
872
|
|
|
'syllabic': 'plus', |
|
873
|
|
|
'voice': 'plus', |
|
874
|
|
|
'nasal': 'minus', |
|
875
|
|
|
'retroflex': 'minus', |
|
876
|
|
|
'lateral': 'minus', |
|
877
|
|
|
'high': 'mid', |
|
878
|
|
|
'back': 'back', |
|
879
|
|
|
'round': 'plus', |
|
880
|
|
|
'long': 'minus', |
|
881
|
|
|
'aspirated': 'minus', |
|
882
|
|
|
}, |
|
883
|
|
|
'ɒ': { |
|
884
|
|
|
'manner': 'low vowel', |
|
885
|
|
|
'syllabic': 'plus', |
|
886
|
|
|
'voice': 'plus', |
|
887
|
|
|
'nasal': 'minus', |
|
888
|
|
|
'retroflex': 'minus', |
|
889
|
|
|
'lateral': 'minus', |
|
890
|
|
|
'high': 'low', |
|
891
|
|
|
'back': 'back', |
|
892
|
|
|
'round': 'minus', |
|
893
|
|
|
'long': 'minus', |
|
894
|
|
|
'aspirated': 'minus', |
|
895
|
|
|
}, |
|
896
|
|
|
'ː': {'long': 'plus', 'supplemental': 'True'}, |
|
897
|
|
|
'ʰ': {'aspirated': 'plus', 'supplemental': 'True'}, |
|
898
|
|
|
} # type: Dict[str, Dict[str, str]] |
|
899
|
|
|
|
|
900
|
|
|
phones_kondrak = { |
|
901
|
|
|
'a': { |
|
902
|
|
|
'place': 'velar', |
|
903
|
|
|
'manner': 'low vowel', |
|
904
|
|
|
'syllabic': 'plus', |
|
905
|
|
|
'voice': 'plus', |
|
906
|
|
|
'nasal': 'minus', |
|
907
|
|
|
'retroflex': 'minus', |
|
908
|
|
|
'lateral': 'minus', |
|
909
|
|
|
'high': 'low', |
|
910
|
1 |
|
'back': 'central', |
|
911
|
|
|
'round': 'minus', |
|
912
|
|
|
}, |
|
913
|
|
|
'b': { |
|
914
|
|
|
'place': 'bilabial', |
|
915
|
|
|
'manner': 'stop', |
|
916
|
|
|
'syllabic': 'minus', |
|
917
|
|
|
'voice': 'plus', |
|
918
|
|
|
'nasal': 'minus', |
|
919
|
|
|
'retroflex': 'minus', |
|
920
|
|
|
'lateral': 'minus', |
|
921
|
|
|
}, |
|
922
|
|
|
'c': { |
|
923
|
|
|
'place': 'alveolar', |
|
924
|
|
|
'manner': 'affricate', |
|
925
|
|
|
'syllabic': 'minus', |
|
926
|
|
|
'voice': 'minus', |
|
927
|
|
|
'nasal': 'minus', |
|
928
|
|
|
'retroflex': 'minus', |
|
929
|
|
|
'lateral': 'minus', |
|
930
|
|
|
}, |
|
931
|
|
|
'd': { |
|
932
|
|
|
'place': 'alveolar', |
|
933
|
|
|
'manner': 'stop', |
|
934
|
|
|
'syllabic': 'minus', |
|
935
|
|
|
'voice': 'plus', |
|
936
|
|
|
'nasal': 'minus', |
|
937
|
|
|
'retroflex': 'minus', |
|
938
|
|
|
'lateral': 'minus', |
|
939
|
|
|
}, |
|
940
|
|
|
'e': { |
|
941
|
|
|
'place': 'palatal', |
|
942
|
|
|
'manner': 'mid vowel', |
|
943
|
|
|
'syllabic': 'plus', |
|
944
|
|
|
'voice': 'plus', |
|
945
|
|
|
'nasal': 'minus', |
|
946
|
|
|
'retroflex': 'minus', |
|
947
|
|
|
'lateral': 'minus', |
|
948
|
|
|
'high': 'mid', |
|
949
|
|
|
'back': 'front', |
|
950
|
|
|
'round': 'minus', |
|
951
|
|
|
}, |
|
952
|
|
|
'f': { |
|
953
|
|
|
'place': 'labiodental', |
|
954
|
|
|
'manner': 'fricative', |
|
955
|
|
|
'syllabic': 'minus', |
|
956
|
|
|
'voice': 'minus', |
|
957
|
|
|
'nasal': 'minus', |
|
958
|
|
|
'retroflex': 'minus', |
|
959
|
|
|
'lateral': 'minus', |
|
960
|
|
|
}, |
|
961
|
|
|
'g': { |
|
962
|
|
|
'place': 'velar', |
|
963
|
|
|
'manner': 'stop', |
|
964
|
|
|
'syllabic': 'minus', |
|
965
|
|
|
'voice': 'plus', |
|
966
|
|
|
'nasal': 'minus', |
|
967
|
|
|
'retroflex': 'minus', |
|
968
|
|
|
'lateral': 'minus', |
|
969
|
|
|
}, |
|
970
|
|
|
'h': { |
|
971
|
|
|
'place': 'glottal', |
|
972
|
|
|
'manner': 'fricative', |
|
973
|
|
|
'syllabic': 'minus', |
|
974
|
|
|
'voice': 'minus', |
|
975
|
|
|
'nasal': 'minus', |
|
976
|
|
|
'retroflex': 'minus', |
|
977
|
|
|
'lateral': 'minus', |
|
978
|
|
|
}, |
|
979
|
|
|
'i': { |
|
980
|
|
|
'place': 'palatal', |
|
981
|
|
|
'manner': 'high vowel', |
|
982
|
|
|
'syllabic': 'plus', |
|
983
|
|
|
'voice': 'plus', |
|
984
|
|
|
'nasal': 'minus', |
|
985
|
|
|
'retroflex': 'minus', |
|
986
|
|
|
'lateral': 'minus', |
|
987
|
|
|
'high': 'high', |
|
988
|
|
|
'back': 'front', |
|
989
|
|
|
'round': 'plus', |
|
990
|
|
|
}, |
|
991
|
|
|
'j': { |
|
992
|
|
|
'place': 'alveolar', |
|
993
|
|
|
'manner': 'affricate', |
|
994
|
|
|
'syllabic': 'minus', |
|
995
|
|
|
'voice': 'plus', |
|
996
|
|
|
'nasal': 'minus', |
|
997
|
|
|
'retroflex': 'minus', |
|
998
|
|
|
'lateral': 'minus', |
|
999
|
|
|
}, |
|
1000
|
|
|
'k': { |
|
1001
|
|
|
'place': 'velar', |
|
1002
|
|
|
'manner': 'stop', |
|
1003
|
|
|
'syllabic': 'minus', |
|
1004
|
|
|
'voice': 'minus', |
|
1005
|
|
|
'nasal': 'minus', |
|
1006
|
|
|
'retroflex': 'minus', |
|
1007
|
|
|
'lateral': 'minus', |
|
1008
|
|
|
}, |
|
1009
|
|
|
'l': { |
|
1010
|
|
|
'place': 'alveolar', |
|
1011
|
|
|
'manner': 'approximant', |
|
1012
|
|
|
'syllabic': 'minus', |
|
1013
|
|
|
'voice': 'plus', |
|
1014
|
|
|
'nasal': 'minus', |
|
1015
|
|
|
'retroflex': 'minus', |
|
1016
|
|
|
'lateral': 'plus', |
|
1017
|
|
|
}, |
|
1018
|
|
|
'm': { |
|
1019
|
|
|
'place': 'bilabial', |
|
1020
|
|
|
'manner': 'stop', |
|
1021
|
|
|
'syllabic': 'minus', |
|
1022
|
|
|
'voice': 'plus', |
|
1023
|
|
|
'nasal': 'plus', |
|
1024
|
|
|
'retroflex': 'minus', |
|
1025
|
|
|
'lateral': 'minus', |
|
1026
|
|
|
}, |
|
1027
|
|
|
'n': { |
|
1028
|
|
|
'place': 'alveolar', |
|
1029
|
|
|
'manner': 'stop', |
|
1030
|
|
|
'syllabic': 'minus', |
|
1031
|
|
|
'voice': 'plus', |
|
1032
|
|
|
'nasal': 'plus', |
|
1033
|
|
|
'retroflex': 'minus', |
|
1034
|
|
|
'lateral': 'minus', |
|
1035
|
|
|
}, |
|
1036
|
|
|
'o': { |
|
1037
|
|
|
'place': 'velar', |
|
1038
|
|
|
'manner': 'mid vowel', |
|
1039
|
|
|
'syllabic': 'plus', |
|
1040
|
|
|
'voice': 'plus', |
|
1041
|
|
|
'nasal': 'minus', |
|
1042
|
|
|
'retroflex': 'minus', |
|
1043
|
|
|
'lateral': 'minus', |
|
1044
|
|
|
'high': 'mid', |
|
1045
|
|
|
'back': 'back', |
|
1046
|
|
|
'round': 'plus', |
|
1047
|
|
|
}, |
|
1048
|
|
|
'p': { |
|
1049
|
|
|
'place': 'bilabial', |
|
1050
|
|
|
'manner': 'stop', |
|
1051
|
|
|
'syllabic': 'minus', |
|
1052
|
|
|
'voice': 'minus', |
|
1053
|
|
|
'nasal': 'minus', |
|
1054
|
|
|
'retroflex': 'minus', |
|
1055
|
|
|
'lateral': 'minus', |
|
1056
|
|
|
}, |
|
1057
|
|
|
'q': { |
|
1058
|
|
|
'place': 'glottal', |
|
1059
|
|
|
'manner': 'stop', |
|
1060
|
|
|
'syllabic': 'minus', |
|
1061
|
|
|
'voice': 'minus', |
|
1062
|
|
|
'nasal': 'minus', |
|
1063
|
|
|
'retroflex': 'minus', |
|
1064
|
|
|
'lateral': 'minus', |
|
1065
|
|
|
}, |
|
1066
|
|
|
'r': { |
|
1067
|
|
|
'place': 'retroflex', |
|
1068
|
|
|
'manner': 'approximant', |
|
1069
|
|
|
'syllabic': 'minus', |
|
1070
|
|
|
'voice': 'plus', |
|
1071
|
|
|
'nasal': 'minus', |
|
1072
|
|
|
'retroflex': 'plus', |
|
1073
|
|
|
'lateral': 'minus', |
|
1074
|
|
|
}, |
|
1075
|
|
|
's': { |
|
1076
|
|
|
'place': 'alveolar', |
|
1077
|
|
|
'manner': 'fricative', |
|
1078
|
|
|
'syllabic': 'minus', |
|
1079
|
|
|
'voice': 'minus', |
|
1080
|
|
|
'nasal': 'minus', |
|
1081
|
|
|
'retroflex': 'minus', |
|
1082
|
|
|
'lateral': 'minus', |
|
1083
|
|
|
}, |
|
1084
|
|
|
't': { |
|
1085
|
|
|
'place': 'alveolar', |
|
1086
|
|
|
'manner': 'stop', |
|
1087
|
|
|
'syllabic': 'minus', |
|
1088
|
|
|
'voice': 'minus', |
|
1089
|
|
|
'nasal': 'minus', |
|
1090
|
|
|
'retroflex': 'minus', |
|
1091
|
|
|
'lateral': 'minus', |
|
1092
|
|
|
}, |
|
1093
|
|
|
'u': { |
|
1094
|
|
|
'place': 'velar', |
|
1095
|
|
|
'manner': 'high vowel', |
|
1096
|
|
|
'syllabic': 'plus', |
|
1097
|
|
|
'voice': 'plus', |
|
1098
|
|
|
'nasal': 'minus', |
|
1099
|
|
|
'retroflex': 'minus', |
|
1100
|
|
|
'lateral': 'minus', |
|
1101
|
|
|
'high': 'high', |
|
1102
|
|
|
'back': 'back', |
|
1103
|
|
|
'round': 'plus', |
|
1104
|
|
|
}, |
|
1105
|
|
|
'v': { |
|
1106
|
|
|
'place': 'labiodental', |
|
1107
|
|
|
'manner': 'fricative', |
|
1108
|
|
|
'syllabic': 'plus', |
|
1109
|
|
|
'voice': 'plus', |
|
1110
|
|
|
'nasal': 'minus', |
|
1111
|
|
|
'retroflex': 'minus', |
|
1112
|
|
|
'lateral': 'minus', |
|
1113
|
|
|
}, |
|
1114
|
|
|
'w': { |
|
1115
|
|
|
'place': 'velar', |
|
1116
|
|
|
'manner': 'high vowel', |
|
1117
|
|
|
'syllabic': 'plus', |
|
1118
|
|
|
'voice': 'plus', |
|
1119
|
|
|
'nasal': 'minus', |
|
1120
|
|
|
'retroflex': 'minus', |
|
1121
|
|
|
'lateral': 'minus', |
|
1122
|
|
|
'high': 'high', |
|
1123
|
|
|
'back': 'back', |
|
1124
|
|
|
'round': 'plus', |
|
1125
|
|
|
'double': 'bilabial', |
|
1126
|
|
|
}, |
|
1127
|
|
|
'x': { |
|
1128
|
|
|
'place': 'velar', |
|
1129
|
|
|
'manner': 'fricative', |
|
1130
|
|
|
'syllabic': 'minus', |
|
1131
|
|
|
'voice': 'minus', |
|
1132
|
|
|
'nasal': 'minus', |
|
1133
|
|
|
'retroflex': 'minus', |
|
1134
|
|
|
'lateral': 'minus', |
|
1135
|
|
|
}, |
|
1136
|
|
|
'y': { |
|
1137
|
|
|
'place': 'velar', |
|
1138
|
|
|
'manner': 'high vowel', |
|
1139
|
|
|
'syllabic': 'plus', |
|
1140
|
|
|
'voice': 'plus', |
|
1141
|
|
|
'nasal': 'minus', |
|
1142
|
|
|
'retroflex': 'minus', |
|
1143
|
|
|
'lateral': 'minus', |
|
1144
|
|
|
'high': 'high', |
|
1145
|
|
|
'back': 'front', |
|
1146
|
|
|
'round': 'minus', |
|
1147
|
|
|
}, |
|
1148
|
|
|
'z': { |
|
1149
|
|
|
'place': 'alveolar', |
|
1150
|
|
|
'manner': 'fricative', |
|
1151
|
|
|
'syllabic': 'minus', |
|
1152
|
|
|
'voice': 'plus', |
|
1153
|
|
|
'nasal': 'minus', |
|
1154
|
|
|
'retroflex': 'minus', |
|
1155
|
|
|
'lateral': 'minus', |
|
1156
|
|
|
}, |
|
1157
|
|
|
'A': {'aspirated': 'plus', 'supplemental': 'True'}, |
|
1158
|
|
|
'B': {'back': 'back', 'supplemental': 'True'}, |
|
1159
|
|
|
'C': {'back': 'central', 'supplemental': 'True'}, |
|
1160
|
|
|
'D': {'place': 'dental', 'supplemental': 'True'}, |
|
1161
|
|
|
'F': {'back': 'front', 'supplemental': 'True'}, |
|
1162
|
|
|
'H': {'long': 'plus', 'supplemental': 'True'}, |
|
1163
|
|
|
'N': {'nasal': 'plus', 'supplemental': 'True'}, |
|
1164
|
|
|
'P': {'place': 'palatal', 'supplemental': 'True'}, |
|
1165
|
|
|
'R': {'round': 'plus', 'supplemental': 'True'}, |
|
1166
|
|
|
'S': {'manner': 'fricative', 'supplemental': 'True'}, |
|
1167
|
|
|
'V': {'place': 'palato-alveolar', 'supplemental': 'True'}, |
|
1168
|
|
|
} # type: Dict[str, Dict[str, str]] |
|
1169
|
|
|
|
|
1170
|
|
|
def __init__( |
|
1171
|
|
|
self, |
|
1172
|
|
|
epsilon: float = 0.0, |
|
1173
|
|
|
c_skip: float = -10, |
|
1174
|
|
|
c_sub: float = 35, |
|
1175
|
|
|
c_exp: float = 45, |
|
1176
|
|
|
c_vwl: float = 10, |
|
1177
|
|
|
mode: str = 'local', |
|
1178
|
|
|
phones: str = 'aline', |
|
1179
|
|
|
normalizer: Callable[[List[float]], float] = max, |
|
1180
|
1 |
|
**kwargs: Any |
|
1181
|
|
|
) -> None: |
|
1182
|
|
|
"""Initialize ALINE instance. |
|
1183
|
|
|
|
|
1184
|
|
|
Parameters |
|
1185
|
|
|
---------- |
|
1186
|
|
|
epsilon : float |
|
1187
|
|
|
The portion (out of 1.0) of the maximum ALINE score, above which |
|
1188
|
|
|
alignments are returned. If set to 0, only the alignments matching |
|
1189
|
|
|
the maximum alignment score are returned. If set to 1, all |
|
1190
|
|
|
alignments scoring 0 or higher are returned. |
|
1191
|
|
|
c_skip : float |
|
1192
|
|
|
The cost of an insertion or deletion |
|
1193
|
|
|
c_sub : float |
|
1194
|
|
|
The cost of a substitution |
|
1195
|
|
|
c_exp : float |
|
1196
|
|
|
The cost of an expansion or contraction |
|
1197
|
|
|
c_vwl : float |
|
1198
|
|
|
The additional cost of a vowel substitution, expansion, or |
|
1199
|
|
|
contraction |
|
1200
|
|
|
mode : str |
|
1201
|
|
|
Alignment mode, which can be ``local`` (default), ``global``, |
|
1202
|
|
|
``half-local``, or ``semi-global`` |
|
1203
|
|
|
phones : str |
|
1204
|
|
|
Phonetic symbol set, which can be: |
|
1205
|
|
|
- ``aline`` selects Kondrak's original symbols set |
|
1206
|
|
|
- ``ipa`` selects IPA symbols |
|
1207
|
|
|
normalizer : function |
|
1208
|
|
|
A function that takes an list and computes a normalization term |
|
1209
|
|
|
by which the edit distance is divided (max by default). For the |
|
1210
|
|
|
normalization proposed by Downey, et al. (2008), set this to: |
|
1211
|
|
|
``lambda x: sum(x)/len(x)`` |
|
1212
|
|
|
**kwargs |
|
1213
|
|
|
Arbitrary keyword arguments |
|
1214
|
|
|
|
|
1215
|
|
|
|
|
1216
|
|
|
.. versionadded:: 0.4.0 |
|
1217
|
|
|
|
|
1218
|
|
|
""" |
|
1219
|
|
|
super(ALINE, self).__init__(**kwargs) |
|
1220
|
|
|
self._epsilon = epsilon |
|
1221
|
|
|
self._c_skip = c_skip |
|
1222
|
|
|
self._c_sub = c_sub |
|
1223
|
|
|
self._c_exp = c_exp |
|
1224
|
|
|
self._c_vwl = c_vwl |
|
1225
|
|
|
self._mode = mode |
|
1226
|
|
|
if self._mode not in {'local', 'global', 'half-local', 'semi-global'}: |
|
1227
|
|
|
self._mode = 'local' |
|
1228
|
|
|
if phones == 'ipa': |
|
1229
|
1 |
|
self._phones = self.phones_ipa |
|
1230
|
1 |
|
else: |
|
1231
|
1 |
|
self._phones = self.phones_kondrak |
|
1232
|
1 |
|
self._normalizer = normalizer |
|
1233
|
1 |
|
|
|
1234
|
1 |
|
def alignment(self, src: str, tar: str) -> Tuple[float, str, str]: |
|
1235
|
1 |
|
"""Return the top ALINE alignment of two strings. |
|
1236
|
1 |
|
|
|
1237
|
1 |
|
The `top` ALINE alignment is the first alignment with the best score. |
|
1238
|
1 |
|
The purpose of this function is to have a single tuple as a return |
|
1239
|
1 |
|
value. |
|
1240
|
|
|
|
|
1241
|
1 |
|
Parameters |
|
1242
|
1 |
|
---------- |
|
1243
|
|
|
src : str |
|
1244
|
1 |
|
Source string for comparison |
|
1245
|
|
|
tar : str |
|
1246
|
|
|
Target string for comparison |
|
1247
|
|
|
|
|
1248
|
|
|
Returns |
|
1249
|
|
|
------- |
|
1250
|
|
|
tuple(float, str, str) |
|
1251
|
|
|
ALINE alignment and its score |
|
1252
|
|
|
|
|
1253
|
|
|
Examples |
|
1254
|
|
|
-------- |
|
1255
|
|
|
>>> cmp = ALINE() |
|
1256
|
|
|
>>> cmp.alignment('cat', 'hat') |
|
1257
|
|
|
(50.0, 'c ‖ a t ‖', 'h ‖ a t ‖') |
|
1258
|
|
|
>>> cmp.alignment('niall', 'neil') |
|
1259
|
|
|
(90.0, '‖ n i a ll ‖', '‖ n e i l ‖') |
|
1260
|
|
|
>>> cmp.alignment('aluminum', 'catalan') |
|
1261
|
|
|
(81.5, '‖ a l u m ‖ inum', 'cat ‖ a l a n ‖') |
|
1262
|
|
|
>>> cmp.alignment('atcg', 'tagc') |
|
1263
|
|
|
(65.0, '‖ a t c ‖ g', 't ‖ a g c ‖') |
|
1264
|
|
|
|
|
1265
|
|
|
|
|
1266
|
|
|
.. versionadded:: 0.4.1 |
|
1267
|
|
|
|
|
1268
|
|
|
""" |
|
1269
|
|
|
return cast(List[Tuple[float, str, str]], self.alignments(src, tar))[0] |
|
1270
|
|
|
|
|
1271
|
|
|
def alignments( |
|
1272
|
|
|
self, src: str, tar: str, score_only: bool = False |
|
1273
|
|
|
) -> Union[float, List[Tuple[float, str, str]]]: |
|
1274
|
|
|
"""Return the ALINE alignments of two strings. |
|
1275
|
|
|
|
|
1276
|
|
|
Parameters |
|
1277
|
|
|
---------- |
|
1278
|
|
|
src : str |
|
1279
|
1 |
|
Source string for comparison |
|
1280
|
|
|
tar : str |
|
1281
|
1 |
|
Target string for comparison |
|
1282
|
|
|
score_only : bool |
|
1283
|
|
|
Return the score only, not the alignments |
|
1284
|
|
|
|
|
1285
|
|
|
Returns |
|
1286
|
|
|
------- |
|
1287
|
|
|
list(tuple(float, str, str) or float |
|
1288
|
|
|
ALINE alignments and their scores or the top score |
|
1289
|
|
|
|
|
1290
|
|
|
Examples |
|
1291
|
|
|
-------- |
|
1292
|
|
|
>>> cmp = ALINE() |
|
1293
|
|
|
>>> cmp.alignments('cat', 'hat') |
|
1294
|
|
|
[(50.0, 'c ‖ a t ‖', 'h ‖ a t ‖')] |
|
1295
|
|
|
>>> cmp.alignments('niall', 'neil') |
|
1296
|
|
|
[(90.0, '‖ n i a ll ‖', '‖ n e i l ‖')] |
|
1297
|
|
|
>>> cmp.alignments('aluminum', 'catalan') |
|
1298
|
|
|
[(81.5, '‖ a l u m ‖ inum', 'cat ‖ a l a n ‖')] |
|
1299
|
|
|
>>> cmp.alignments('atcg', 'tagc') |
|
1300
|
|
|
[(65.0, '‖ a t c ‖ g', 't ‖ a g c ‖'), (65.0, 'a ‖ tc - g ‖', |
|
1301
|
|
|
'‖ t a g ‖ c')] |
|
1302
|
|
|
|
|
1303
|
|
|
|
|
1304
|
|
|
.. versionadded:: 0.4.0 |
|
1305
|
|
|
.. versionchanged:: 0.4.1 |
|
1306
|
|
|
Renamed from .alignment to .alignments |
|
1307
|
|
|
|
|
1308
|
|
|
""" |
|
1309
|
|
|
|
|
1310
|
|
|
def _sig_skip(*args: Any) -> float: |
|
1311
|
|
|
return self._c_skip |
|
1312
|
|
|
|
|
1313
|
|
|
def _sig_sub(seg1: Dict[str, float], seg2: Dict[str, float]) -> float: |
|
1314
|
|
|
return ( |
|
1315
|
|
|
self._c_sub |
|
1316
|
|
|
- _delta(seg1, seg2) |
|
1317
|
|
|
- _sig_vwl(seg1) |
|
1318
|
1 |
|
- _sig_vwl(seg2) |
|
1319
|
1 |
|
) |
|
1320
|
|
|
|
|
1321
|
1 |
|
def _sig_exp( |
|
1322
|
1 |
|
seg1: Dict[str, float], |
|
1323
|
|
|
seg2a: Dict[str, float], |
|
1324
|
|
|
seg2b: Dict[str, float], |
|
1325
|
|
|
) -> float: |
|
1326
|
|
|
return ( |
|
1327
|
|
|
self._c_exp |
|
1328
|
|
|
- _delta(seg1, seg2a) |
|
1329
|
1 |
|
- _delta(seg1, seg2b) |
|
1330
|
1 |
|
- _sig_vwl(seg1) |
|
1331
|
|
|
- max(_sig_vwl(seg2a), _sig_vwl(seg2b)) |
|
1332
|
|
|
) |
|
1333
|
|
|
|
|
1334
|
|
|
def _sig_vwl(seg1: Dict[str, float]) -> float: |
|
1335
|
|
|
return ( |
|
1336
|
|
|
0.0 |
|
1337
|
|
|
if seg1['manner'] > self.feature_weights['high vowel'] |
|
1338
|
1 |
|
else self._c_vwl |
|
1339
|
1 |
|
) |
|
1340
|
|
|
|
|
1341
|
|
|
def _delta(seg1: Dict[str, float], seg2: Dict[str, float]) -> float: |
|
1342
|
|
|
features = ( |
|
1343
|
|
|
self.c_features |
|
1344
|
|
|
if max(seg1['manner'], seg2['manner']) |
|
1345
|
1 |
|
> self.feature_weights['high vowel'] |
|
1346
|
1 |
|
else self.v_features |
|
1347
|
|
|
) |
|
1348
|
|
|
diff = 0.0 |
|
1349
|
|
|
for f in features: |
|
1350
|
|
|
diff += ( |
|
1351
|
|
|
abs(seg1.get(f, 0.0) - seg2.get(f, 0.0)) * self.salience[f] |
|
1352
|
1 |
|
) |
|
1353
|
1 |
|
return diff |
|
1354
|
1 |
|
|
|
1355
|
|
|
def _retrieve( |
|
1356
|
|
|
i: int, j: int, score: float, out: List[Tuple[str, str]] |
|
1357
|
1 |
|
) -> None: |
|
1358
|
|
|
def _record(score: float, out: List[Tuple[str, str]]) -> None: |
|
1359
|
1 |
|
out.append(('‖', '‖')) |
|
1360
|
1 |
|
for i1 in range(i - 1, -1, -1): |
|
1361
|
1 |
|
out.append((src_tok[i1], '')) |
|
1362
|
1 |
|
for j1 in range(j - 1, -1, -1): |
|
1363
|
1 |
|
out.append(('', tar_tok[j1])) |
|
1364
|
1 |
|
if self._mode == 'global': |
|
1365
|
1 |
|
score += (i + j) * _sig_skip('') |
|
1366
|
1 |
|
|
|
1367
|
1 |
|
out = out[::-1] |
|
1368
|
|
|
|
|
1369
|
1 |
|
src_alignment = [] |
|
1370
|
|
|
tar_alignment = [] |
|
1371
|
1 |
|
|
|
1372
|
1 |
|
out.append(('‖', '‖')) |
|
1373
|
|
|
part = 0 |
|
1374
|
1 |
|
s_segment = '' # type: Union[str, List[str]] |
|
1375
|
1 |
|
t_segment = '' # type: Union[str, List[str]] |
|
1376
|
1 |
|
for ss, ts in out: |
|
1377
|
1 |
|
if ss == '‖': |
|
1378
|
1 |
|
if part % 2 == 0: |
|
1379
|
1 |
|
src_alignment.append(s_segment) |
|
1380
|
1 |
|
tar_alignment.append(t_segment) |
|
1381
|
1 |
|
s_segment = [] |
|
1382
|
1 |
|
t_segment = [] |
|
1383
|
1 |
|
else: |
|
1384
|
1 |
|
src_alignment.append(' '.join(s_segment)) |
|
1385
|
|
|
tar_alignment.append(' '.join(t_segment)) |
|
1386
|
1 |
|
s_segment = '' |
|
1387
|
1 |
|
t_segment = '' |
|
1388
|
1 |
|
part += 1 |
|
1389
|
1 |
|
else: |
|
1390
|
1 |
|
if part % 2 == 0: |
|
1391
|
|
|
s_segment = cast(str, s_segment) + ss |
|
1392
|
1 |
|
t_segment = cast(str, t_segment) + ts |
|
1393
|
1 |
|
else: |
|
1394
|
1 |
|
cast(List[str], s_segment).append( |
|
1395
|
|
|
ss + ' ' * (len(ts) - len(ss)) |
|
1396
|
1 |
|
) |
|
1397
|
1 |
|
cast(List[str], t_segment).append( |
|
1398
|
|
|
ts + ' ' * (len(ss) - len(ts)) |
|
1399
|
1 |
|
) |
|
1400
|
1 |
|
|
|
1401
|
|
|
src_alignment_str = ' ‖ '.join( |
|
1402
|
1 |
|
cast(List[str], src_alignment) |
|
1403
|
1 |
|
).strip() |
|
1404
|
|
|
tar_alignment_str = ' ‖ '.join( |
|
1405
|
1 |
|
cast(List[str], tar_alignment) |
|
1406
|
1 |
|
).strip() |
|
1407
|
1 |
|
|
|
1408
|
|
|
alignments.append( |
|
1409
|
1 |
|
(score, src_alignment_str, tar_alignment_str) |
|
1410
|
|
|
) |
|
1411
|
|
|
return |
|
1412
|
|
|
|
|
1413
|
|
|
if s_mat[i, j] == 0: |
|
1414
|
|
|
_record(score, out) |
|
1415
|
|
|
return |
|
1416
|
|
|
else: |
|
1417
|
1 |
|
if ( |
|
1418
|
1 |
|
i > 0 |
|
1419
|
|
|
and j > 0 |
|
1420
|
|
|
and s_mat[i - 1, j - 1] |
|
1421
|
1 |
|
+ _sig_sub(src_feat_wt[i - 1], tar_feat_wt[j - 1]) |
|
1422
|
|
|
+ score |
|
1423
|
|
|
>= threshold |
|
1424
|
|
|
): |
|
1425
|
|
|
loc_out = deepcopy(out) |
|
1426
|
|
|
loc_out.append((src_tok[i - 1], tar_tok[j - 1])) |
|
1427
|
1 |
|
_retrieve( |
|
1428
|
|
|
i - 1, |
|
1429
|
1 |
|
j - 1, |
|
1430
|
|
|
score |
|
1431
|
|
|
+ _sig_sub(src_feat_wt[i - 1], tar_feat_wt[j - 1]), |
|
1432
|
|
|
loc_out, |
|
1433
|
|
|
) |
|
1434
|
1 |
|
loc_out.pop() |
|
1435
|
1 |
|
|
|
1436
|
1 |
View Code Duplication |
if ( |
|
|
|
|
|
|
1437
|
1 |
|
j > 0 |
|
1438
|
|
|
and s_mat[i, j - 1] + _sig_skip(tar_tok[j - 1]) + score |
|
1439
|
1 |
|
>= threshold |
|
1440
|
|
|
): |
|
1441
|
|
|
loc_out = deepcopy(out) |
|
1442
|
|
|
loc_out.append(('-', tar_tok[j - 1])) |
|
1443
|
|
|
_retrieve( |
|
1444
|
|
|
i, j - 1, score + _sig_skip(tar_tok[j - 1]), loc_out |
|
1445
|
|
|
) |
|
1446
|
|
|
loc_out.pop() |
|
1447
|
1 |
|
|
|
1448
|
1 |
View Code Duplication |
if ( |
|
|
|
|
|
|
1449
|
|
|
i > 0 |
|
1450
|
|
|
and j > 1 |
|
1451
|
|
|
and s_mat[i - 1, j - 2] |
|
1452
|
|
|
+ _sig_exp( |
|
1453
|
|
|
src_feat_wt[i - 1], |
|
1454
|
1 |
|
tar_feat_wt[j - 2], |
|
1455
|
|
|
tar_feat_wt[j - 1], |
|
1456
|
|
|
) |
|
1457
|
|
|
+ score |
|
1458
|
|
|
>= threshold |
|
1459
|
|
|
): |
|
1460
|
1 |
|
loc_out = deepcopy(out) |
|
1461
|
|
|
loc_out.append( |
|
1462
|
1 |
|
(src_tok[i - 1], tar_tok[j - 2] + tar_tok[j - 1],) |
|
1463
|
|
|
) |
|
1464
|
|
|
_retrieve( |
|
1465
|
|
|
i - 1, |
|
1466
|
|
|
j - 2, |
|
1467
|
1 |
|
score |
|
1468
|
1 |
|
+ _sig_exp( |
|
1469
|
1 |
|
src_feat_wt[i - 1], |
|
1470
|
1 |
|
tar_feat_wt[j - 2], |
|
1471
|
|
|
tar_feat_wt[j - 1], |
|
1472
|
1 |
|
), |
|
1473
|
|
|
loc_out, |
|
1474
|
|
|
) |
|
1475
|
|
|
loc_out.pop() |
|
1476
|
|
|
|
|
1477
|
|
View Code Duplication |
if ( |
|
|
|
|
|
|
1478
|
|
|
i > 0 |
|
1479
|
|
|
and s_mat[i - 1, j] + _sig_skip(src_tok[i - 1]) + score |
|
1480
|
1 |
|
>= threshold |
|
1481
|
1 |
|
): |
|
1482
|
|
|
loc_out = deepcopy(out) |
|
1483
|
|
|
loc_out.append((src_tok[i - 1], '-')) |
|
1484
|
|
|
_retrieve( |
|
1485
|
|
|
i - 1, j, score + _sig_skip(src_tok[i - 1]), loc_out |
|
1486
|
|
|
) |
|
1487
|
1 |
|
loc_out.pop() |
|
1488
|
|
|
|
|
1489
|
|
View Code Duplication |
if ( |
|
|
|
|
|
|
1490
|
|
|
i > 1 |
|
1491
|
|
|
and j > 0 |
|
1492
|
|
|
and s_mat[i - 2, j - 1] |
|
1493
|
1 |
|
+ _sig_exp( |
|
1494
|
|
|
tar_feat_wt[j - 1], |
|
1495
|
1 |
|
src_feat_wt[i - 2], |
|
1496
|
|
|
src_feat_wt[i - 1], |
|
1497
|
1 |
|
) |
|
1498
|
1 |
|
+ score |
|
1499
|
|
|
>= threshold |
|
1500
|
1 |
|
): |
|
1501
|
1 |
|
loc_out = deepcopy(out) |
|
1502
|
1 |
|
loc_out.append( |
|
1503
|
1 |
|
(src_tok[i - 2] + src_tok[i - 1], tar_tok[j - 1],) |
|
1504
|
1 |
|
) |
|
1505
|
1 |
|
_retrieve( |
|
1506
|
1 |
|
i - 2, |
|
1507
|
1 |
|
j - 1, |
|
1508
|
1 |
|
score |
|
1509
|
1 |
|
+ _sig_exp( |
|
1510
|
|
|
tar_feat_wt[j - 1], |
|
1511
|
1 |
|
src_feat_wt[i - 2], |
|
1512
|
1 |
|
src_feat_wt[i - 1], |
|
1513
|
|
|
), |
|
1514
|
1 |
|
loc_out, |
|
1515
|
1 |
|
) |
|
1516
|
1 |
|
loc_out.pop() |
|
1517
|
1 |
|
|
|
1518
|
1 |
|
sg_max = 0.0 |
|
1519
|
1 |
|
|
|
1520
|
1 |
|
src_tok = [] # type: List[str] |
|
1521
|
1 |
|
src_feat = [] # type: List[Dict[str, str]] |
|
1522
|
1 |
|
tar_tok = [] # type: List[str] |
|
1523
|
|
|
tar_feat = [] # type: List[Dict[str, str]] |
|
1524
|
1 |
|
|
|
1525
|
1 |
|
for ch in src: |
|
1526
|
1 |
|
if ch in self._phones: |
|
1527
|
1 |
|
src_tok.append(ch) |
|
1528
|
|
|
src_feat.append(dict(self._phones[ch])) |
|
1529
|
1 |
|
for ch in tar: |
|
1530
|
1 |
|
if ch in self._phones: |
|
1531
|
1 |
|
tar_tok.append(ch) |
|
1532
|
1 |
|
tar_feat.append(dict(self._phones[ch])) |
|
1533
|
1 |
|
|
|
1534
|
1 |
|
for i in range(1, len(src_feat)): |
|
1535
|
1 |
|
if 'supplemental' in src_feat[i]: |
|
1536
|
1 |
|
j = i - 1 |
|
1537
|
1 |
|
while j > -1: |
|
1538
|
|
|
if 'supplemental' not in src_feat[j]: |
|
1539
|
1 |
|
src_tok[j] += src_tok[i] |
|
1540
|
1 |
|
for key, value in src_feat[i].items(): |
|
1541
|
1 |
|
if key != 'supplemental': |
|
1542
|
1 |
|
src_feat[j][key] = value |
|
1543
|
|
|
j = 0 |
|
1544
|
1 |
|
j -= 1 |
|
1545
|
1 |
|
|
|
1546
|
1 |
|
zipped = [ |
|
1547
|
1 |
|
fb for fb in zip(src_feat, src_tok) if 'supplemental' not in fb[0] |
|
1548
|
1 |
|
] |
|
1549
|
1 |
|
if zipped: |
|
1550
|
1 |
|
src_feat, src_tok = zip(*zipped) # type: ignore |
|
1551
|
1 |
|
else: |
|
1552
|
|
|
src_feat, src_tok = [], [] |
|
1553
|
1 |
|
|
|
1554
|
1 |
|
src_feat_wt = [] # type: List[Dict[str, float]] |
|
1555
|
|
|
for f_dict in src_feat: |
|
1556
|
1 |
|
src_feat_wt.append( |
|
1557
|
|
|
{ |
|
1558
|
1 |
|
key: self.feature_weights[f_dict[key]] |
|
1559
|
1 |
|
for key in f_dict.keys() |
|
1560
|
1 |
|
} |
|
1561
|
1 |
|
) |
|
1562
|
1 |
|
|
|
1563
|
|
|
src_len = len(src_tok) |
|
1564
|
1 |
|
|
|
1565
|
1 |
|
for i in range(1, len(tar_feat)): |
|
1566
|
1 |
|
if 'supplemental' in tar_feat[i]: |
|
1567
|
|
|
j = i - 1 |
|
1568
|
|
|
while j > -1: |
|
1569
|
|
|
if 'supplemental' not in tar_feat[j]: |
|
1570
|
|
|
tar_tok[j] += tar_tok[i] |
|
1571
|
|
|
for key, value in tar_feat[i].items(): |
|
1572
|
|
|
if key != 'supplemental': |
|
1573
|
|
|
tar_feat[j][key] = value |
|
1574
|
|
|
j = 0 |
|
1575
|
|
|
j -= 1 |
|
1576
|
|
|
|
|
1577
|
|
|
zipped = [ |
|
1578
|
|
|
fb for fb in zip(tar_feat, tar_tok) if 'supplemental' not in fb[0] |
|
1579
|
|
|
] |
|
1580
|
|
|
if zipped: |
|
1581
|
1 |
|
tar_feat, tar_tok = zip(*zipped) # type: ignore |
|
1582
|
1 |
|
else: |
|
1583
|
1 |
|
tar_feat, tar_tok = [], [] |
|
1584
|
1 |
|
|
|
1585
|
|
|
tar_feat_wt = [] # type: List[Dict[str, float]] |
|
1586
|
1 |
|
for f_dict in tar_feat: |
|
1587
|
|
|
tar_feat_wt.append( |
|
1588
|
1 |
|
{ |
|
1589
|
1 |
|
key: self.feature_weights[f_dict[key]] |
|
1590
|
|
|
for key in f_dict.keys() |
|
1591
|
1 |
|
} |
|
1592
|
|
|
) |
|
1593
|
1 |
|
|
|
1594
|
1 |
|
tar_len = len(tar_tok) |
|
1595
|
|
|
|
|
1596
|
1 |
|
s_mat = zeros((src_len + 1, tar_len + 1), dtype=float_) |
|
1597
|
|
|
|
|
1598
|
1 |
|
if self._mode == 'global': |
|
1599
|
|
|
for i in range(1, src_len + 1): |
|
1600
|
1 |
|
s_mat[i, 0] = s_mat[i - 1, 0] + _sig_skip(src_tok[i - 1]) |
|
1601
|
1 |
|
for j in range(1, tar_len + 1): |
|
1602
|
1 |
|
s_mat[0, j] = s_mat[0, j - 1] + _sig_skip(tar_tok[j - 1]) |
|
1603
|
|
|
|
|
1604
|
|
|
for i in range(1, src_len + 1): |
|
1605
|
1 |
|
for j in range(1, tar_len + 1): |
|
1606
|
1 |
|
s_mat[i, j] = max( |
|
1607
|
|
|
s_mat[i - 1, j] + _sig_skip(src_feat_wt[i - 1]), |
|
1608
|
|
|
s_mat[i, j - 1] + _sig_skip(tar_feat_wt[j - 1]), |
|
1609
|
1 |
|
s_mat[i - 1, j - 1] |
|
1610
|
1 |
|
+ _sig_sub(src_feat_wt[i - 1], tar_feat_wt[j - 1]), |
|
1611
|
1 |
|
s_mat[i - 1, j - 2] |
|
1612
|
1 |
|
+ _sig_exp( |
|
1613
|
1 |
|
src_feat_wt[i - 1], |
|
1614
|
1 |
|
tar_feat_wt[j - 2], |
|
1615
|
1 |
|
tar_feat_wt[j - 1], |
|
1616
|
1 |
|
) |
|
1617
|
1 |
|
if j > 1 |
|
1618
|
|
|
else -inf, |
|
1619
|
1 |
|
s_mat[i - 2, j - 1] |
|
1620
|
1 |
|
+ _sig_exp( |
|
1621
|
|
|
tar_feat_wt[j - 1], |
|
1622
|
1 |
|
src_feat_wt[i - 2], |
|
1623
|
|
|
src_feat_wt[i - 1], |
|
1624
|
1 |
|
) |
|
1625
|
|
|
if i > 1 |
|
1626
|
|
|
else -inf, |
|
1627
|
|
|
0 if self._mode in {'local', 'half-local'} else -inf, |
|
1628
|
|
|
) |
|
1629
|
|
|
|
|
1630
|
|
|
if s_mat[i, j] > sg_max: |
|
1631
|
|
|
if self._mode == 'semi-global': |
|
1632
|
|
|
if i == src_len or j == tar_len: |
|
1633
|
|
|
sg_max = s_mat[i, j] |
|
1634
|
|
|
else: |
|
1635
|
|
|
sg_max = s_mat[i, j] |
|
1636
|
|
|
|
|
1637
|
|
|
if self._mode in {'global', 'half-local'}: |
|
1638
|
|
|
dp_score = s_mat[src_len, tar_len] |
|
1639
|
|
|
else: |
|
1640
|
|
|
dp_score = s_mat.max() |
|
1641
|
|
|
|
|
1642
|
|
|
if score_only: |
|
1643
|
|
|
return cast(float, dp_score) |
|
1644
|
|
|
|
|
1645
|
|
|
threshold = (1 - self._epsilon) * dp_score |
|
1646
|
|
|
|
|
1647
|
|
|
alignments = [] # type: List[Tuple[float, str, str]] |
|
1648
|
|
|
|
|
1649
|
|
|
for i in range(1, src_len + 1): |
|
1650
|
|
|
for j in range(1, tar_len + 1): |
|
1651
|
|
|
if self._mode in {'global', 'half-local'} and ( |
|
1652
|
|
|
i < src_len or j < tar_len |
|
1653
|
|
|
): |
|
1654
|
|
|
continue |
|
1655
|
1 |
|
if self._mode == 'semi-global' and ( |
|
1656
|
1 |
|
i < src_len and j < tar_len |
|
1657
|
1 |
|
): |
|
1658
|
|
|
continue |
|
1659
|
1 |
|
if s_mat[i, j] >= threshold: |
|
1660
|
|
|
out = [] |
|
1661
|
|
|
for j1 in range(tar_len - 1, j - 1, -1): |
|
1662
|
|
|
out.append(('', tar_tok[j1])) |
|
1663
|
|
|
for i1 in range(src_len - 1, i - 1, -1): |
|
1664
|
|
|
out.append((src_tok[i1], '')) |
|
1665
|
|
|
out.append(('‖', '‖')) |
|
1666
|
|
|
_retrieve(i, j, 0, out) |
|
1667
|
|
|
|
|
1668
|
|
|
return sorted(alignments, key=lambda _: _[0], reverse=True) |
|
1669
|
|
|
|
|
1670
|
|
|
def sim_score(self, src: str, tar: str) -> float: |
|
1671
|
|
|
"""Return the ALINE alignment score of two strings. |
|
1672
|
|
|
|
|
1673
|
|
|
Parameters |
|
1674
|
|
|
---------- |
|
1675
|
|
|
src : str |
|
1676
|
|
|
Source string for comparison |
|
1677
|
|
|
tar : str |
|
1678
|
|
|
Target string for comparison |
|
1679
|
|
|
|
|
1680
|
|
|
Returns |
|
1681
|
|
|
------- |
|
1682
|
|
|
float |
|
1683
|
|
|
ALINE alignment score |
|
1684
|
|
|
|
|
1685
|
|
|
Examples |
|
1686
|
|
|
-------- |
|
1687
|
|
|
>>> cmp = ALINE() |
|
1688
|
|
|
>>> cmp.sim_score('cat', 'hat') |
|
1689
|
|
|
50.0 |
|
1690
|
1 |
|
>>> cmp.sim_score('niall', 'neil') |
|
1691
|
1 |
|
90.0 |
|
1692
|
1 |
|
>>> cmp.sim_score('aluminum', 'catalan') |
|
1693
|
|
|
81.5 |
|
1694
|
|
|
>>> cmp.sim_score('atcg', 'tagc') |
|
1695
|
1 |
|
65.0 |
|
1696
|
|
|
|
|
1697
|
|
|
|
|
1698
|
|
|
.. versionadded:: 0.4.0 |
|
1699
|
|
|
|
|
1700
|
|
|
""" |
|
1701
|
|
|
if src == '' and tar == '': |
|
1702
|
|
|
return 1.0 |
|
1703
|
|
|
return cast(float, self.alignments(src, tar, score_only=True)) |
|
1704
|
|
|
|
|
1705
|
|
|
def sim(self, src: str, tar: str) -> float: |
|
1706
|
|
|
"""Return the normalized ALINE similarity of two strings. |
|
1707
|
|
|
|
|
1708
|
|
|
Parameters |
|
1709
|
|
|
---------- |
|
1710
|
|
|
src : str |
|
1711
|
|
|
Source string for comparison |
|
1712
|
|
|
tar : str |
|
1713
|
|
|
Target string for comparison |
|
1714
|
|
|
|
|
1715
|
|
|
Returns |
|
1716
|
|
|
------- |
|
1717
|
|
|
float |
|
1718
|
|
|
Normalized ALINE similarity |
|
1719
|
|
|
|
|
1720
|
|
|
Examples |
|
1721
|
|
|
-------- |
|
1722
|
|
|
>>> cmp = ALINE() |
|
1723
|
|
|
>>> cmp.dist('cat', 'hat') |
|
1724
|
|
|
0.4117647058823529 |
|
1725
|
|
|
>>> cmp.dist('niall', 'neil') |
|
1726
|
|
|
0.33333333333333337 |
|
1727
|
|
|
>>> cmp.dist('aluminum', 'catalan') |
|
1728
|
|
|
0.5925 |
|
1729
|
|
|
>>> cmp.dist('atcg', 'tagc') |
|
1730
|
|
|
0.45833333333333337 |
|
1731
|
|
|
|
|
1732
|
|
|
|
|
1733
|
|
|
.. versionadded:: 0.4.0 |
|
1734
|
|
|
|
|
1735
|
|
|
""" |
|
1736
|
|
|
num = self.sim_score(src, tar) |
|
1737
|
|
|
if num: |
|
1738
|
|
|
return num / self._normalizer( |
|
1739
|
|
|
[self.sim_score(src, src), self.sim_score(tar, tar)] |
|
1740
|
|
|
) |
|
1741
|
|
|
return 0.0 |
|
1742
|
|
|
|
|
1743
|
|
|
|
|
1744
|
|
|
if __name__ == '__main__': |
|
1745
|
|
|
import doctest |
|
1746
|
|
|
|
|
1747
|
|
|
doctest.testmod(optionflags=doctest.NORMALIZE_WHITESPACE) |
|
1748
|
|
|
|
chrislit /
abydos
Loading history...