asahi417 commited on
Commit
c76e8c8
1 Parent(s): 268df96
Files changed (4) hide show
  1. download_s2t_metadata.py +1 -1
  2. main_s2s.sh +226 -73
  3. main_s2t.sh +18 -1
  4. speaker_embedding_clap.py +35 -0
download_s2t_metadata.py CHANGED
@@ -19,7 +19,7 @@ filename = os.path.basename(url)
19
  subprocess.run(["wget", url, "-O", filename])
20
  df = pd.read_csv(filename, sep='\t', header=None, dtype=str)
21
  df.columns = ["cc_warc", "cc_sha", "cc_document_url", "cc_lineno", "paragraph_digest", "sentence_digest", "text_lid_score", "laser_score", "direction", "side", "line_no"]
22
- df = df[df.side == "jpn"]
23
  df["cc_lineno"] = df["cc_lineno"].astype(int)
24
  df.sort_values(by=["cc_warc", "cc_sha", "cc_document_url", "cc_lineno"], inplace=True)
25
  batch_size = int(len(df)/chunk_size)
 
19
  subprocess.run(["wget", url, "-O", filename])
20
  df = pd.read_csv(filename, sep='\t', header=None, dtype=str)
21
  df.columns = ["cc_warc", "cc_sha", "cc_document_url", "cc_lineno", "paragraph_digest", "sentence_digest", "text_lid_score", "laser_score", "direction", "side", "line_no"]
22
+ df = df[df.side == direction_text]
23
  df["cc_lineno"] = df["cc_lineno"].astype(int)
24
  df.sort_values(by=["cc_warc", "cc_sha", "cc_document_url", "cc_lineno"], inplace=True)
25
  batch_size = int(len(df)/chunk_size)
main_s2s.sh CHANGED
@@ -1,234 +1,387 @@
1
  ####################
2
- # enA-jaA: 718_606 #
3
  ####################
4
  # test
5
  export DATASET_ID=test
6
- export DIRECTION="enA-jaA"
7
  export LINE_NO_START=0
8
  export LINE_NO_END=10
9
  python fetch_dataset_s2s.py
10
  # main
11
- for i in $(seq 1 144);
12
  do
13
  export N_POOL=15
14
  export DATASET_ID=${i}
15
- export DIRECTION="enA-jaA"
 
 
 
 
 
 
 
 
 
 
16
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
17
  export LINE_NO_END=$((DATASET_ID * 2500))
18
  echo ${LINE_NO_START}
19
  python fetch_dataset_s2s.py
20
  done
21
  # tokenize
22
- for i in $(seq 1 144);
23
  do
 
24
  export DATASET_ID=${i}
25
- export DIRECTION="enA-jaA"
 
 
 
 
 
 
 
 
 
 
 
 
 
26
  python tokenize_dataset_s2s.py
27
  done
28
  # speaker embedding
29
- for i in $(seq 1 144);
30
  do
31
  export DATASET_ID=${i}
32
- export DIRECTION="enA-jaA"
 
 
 
 
 
 
33
  python attach_speaker_embedding_s2s.py
34
  done
35
 
36
- ####################
37
- # enA-hiA: 454_942 #
38
- ####################
39
  # test
40
  export DATASET_ID=test
41
- export DIRECTION="enA-hiA"
42
  export LINE_NO_START=0
43
  export LINE_NO_END=10
44
  python fetch_dataset_s2s.py
45
  # main
46
- for i in $(seq 1 91);
47
  do
48
  export N_POOL=15
49
  export DATASET_ID=${i}
50
- export DIRECTION="enA-hiA"
51
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
52
  export LINE_NO_END=$((DATASET_ID * 2500))
53
  echo ${LINE_NO_START}
54
  python fetch_dataset_s2s.py
55
  done
56
  # tokenize
57
- for i in $(seq 1 91);
58
  do
 
59
  export DATASET_ID=${i}
60
- export DIRECTION="enA-hiA"
 
 
 
61
  python tokenize_dataset_s2s.py
62
  done
63
  # speaker embedding
64
- for i in $(seq 1 91);
65
  do
66
  export DATASET_ID=${i}
67
- export DIRECTION="enA-hiA"
68
  python attach_speaker_embedding_s2s.py
69
  done
70
-
71
- ####################
72
- # enA-viA: 740_598 #
73
- ####################
74
  # test
75
  export DATASET_ID=test
76
- export DIRECTION="enA-viA"
77
  export LINE_NO_START=0
78
  export LINE_NO_END=10
79
  python fetch_dataset_s2s.py
80
  # main
81
- for i in $(seq 1 149);
82
  do
83
  export N_POOL=15
84
  export DATASET_ID=${i}
85
- export DIRECTION="enA-viA"
 
 
 
 
 
 
 
 
 
 
86
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
87
  export LINE_NO_END=$((DATASET_ID * 2500))
88
  echo ${LINE_NO_START}
89
  python fetch_dataset_s2s.py
90
  done
91
  # tokenize
92
- for i in $(seq 141 149);
93
  do
 
94
  export DATASET_ID=${i}
95
- export DIRECTION="enA-viA"
 
 
 
96
  python tokenize_dataset_s2s.py
97
  done
98
- for i in 25 26 33 34 35 36 37 38 39 40;
99
  do
 
100
  export DATASET_ID=${i}
101
- export DIRECTION="enA-viA"
 
 
 
102
  python tokenize_dataset_s2s.py
103
  done
104
  # speaker embedding
105
- for i in $(seq 1 144);
106
  do
107
  export DATASET_ID=${i}
108
- export DIRECTION="enA-viA"
109
  python attach_speaker_embedding_s2s.py
110
  done
111
-
112
- ####################
113
- # enA-koA: 511_358 #
114
- ####################
 
 
 
 
 
 
115
  # test
116
  export DATASET_ID=test
117
- export DIRECTION="enA-koA"
118
  export LINE_NO_START=0
119
  export LINE_NO_END=10
120
  python fetch_dataset_s2s.py
121
  # main
122
- for i in $(seq 1 50);
123
  do
124
  export N_POOL=15
125
  export DATASET_ID=${i}
126
- export DIRECTION="enA-koA"
127
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
128
  export LINE_NO_END=$((DATASET_ID * 2500))
129
  echo ${LINE_NO_START}
130
  python fetch_dataset_s2s.py
131
  done
132
- for i in $(seq 51 100);
133
  do
134
  export N_POOL=15
135
  export DATASET_ID=${i}
136
- export DIRECTION="enA-koA"
137
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
138
  export LINE_NO_END=$((DATASET_ID * 2500))
139
  echo ${LINE_NO_START}
140
  python fetch_dataset_s2s.py
141
  done
142
-
143
- ######################
144
- # enA-zhA: 1_289_192 #
145
- ######################
146
- # test
147
- export DATASET_ID=test
148
- export DIRECTION="enA-zhA"
149
- export LINE_NO_START=0
150
- export LINE_NO_END=10
151
- python fetch_dataset_s2s.py
152
- # main
153
- for i in $(seq 1 258);
154
  do
155
  export N_POOL=15
156
  export DATASET_ID=${i}
157
- export DIRECTION="enA-zhA"
158
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
159
  export LINE_NO_END=$((DATASET_ID * 2500))
160
  echo ${LINE_NO_START}
161
- python fetch_dataset_s2s.py
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
162
  done
163
-
164
  ######################
165
- # enA-frA: 3_054_258 #
166
  ######################
167
  # test
168
  export DATASET_ID=test
169
- export DIRECTION="enA-frA"
170
  export LINE_NO_START=0
171
  export LINE_NO_END=10
172
  python fetch_dataset_s2s.py
173
  # main
174
- for i in $(seq 1 300);
175
  do
176
  export N_POOL=15
177
  export DATASET_ID=${i}
178
- export DIRECTION="enA-frA"
179
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
180
  export LINE_NO_END=$((DATASET_ID * 2500))
181
  echo ${LINE_NO_START}
182
  python fetch_dataset_s2s.py
183
  done
184
- for i in $(seq 300 600);
185
  do
186
  export N_POOL=15
187
  export DATASET_ID=${i}
188
- export DIRECTION="enA-frA"
189
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
190
  export LINE_NO_END=$((DATASET_ID * 2500))
191
  echo ${LINE_NO_START}
192
  python fetch_dataset_s2s.py
193
  done
194
 
195
- ######################
196
- # enA-esA: 2_658_022 #
197
- ######################
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
198
  # test
199
  export DATASET_ID=test
200
- export DIRECTION="enA-esA"
201
  export LINE_NO_START=0
202
  export LINE_NO_END=10
203
  python fetch_dataset_s2s.py
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
204
 
205
- ######################
206
- # enA-deA: 1_965_186 #
207
- ######################
208
  # test
209
  export DATASET_ID=test
210
- export DIRECTION="deA-enA"
211
  export LINE_NO_START=0
212
  export LINE_NO_END=10
213
  python fetch_dataset_s2s.py
214
  # main
215
- for i in $(seq 1 200);
216
  do
217
  export N_POOL=15
218
  export DATASET_ID=${i}
219
- export DIRECTION="deA-enA"
220
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
221
  export LINE_NO_END=$((DATASET_ID * 2500))
222
  echo ${LINE_NO_START}
223
  python fetch_dataset_s2s.py
224
  done
225
- for i in $(seq 201 394);
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
226
  do
227
  export N_POOL=15
228
  export DATASET_ID=${i}
229
- export DIRECTION="deA-enA"
230
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
231
  export LINE_NO_END=$((DATASET_ID * 2500))
232
  echo ${LINE_NO_START}
233
  python fetch_dataset_s2s.py
 
 
 
 
 
 
 
 
 
 
 
 
 
 
234
  done
 
1
  ####################
2
+ # enA-koA: 511_358 #
3
  ####################
4
  # test
5
  export DATASET_ID=test
6
+ export DIRECTION="enA-koA"
7
  export LINE_NO_START=0
8
  export LINE_NO_END=10
9
  python fetch_dataset_s2s.py
10
  # main
11
+ for i in $(seq 1 50);
12
  do
13
  export N_POOL=15
14
  export DATASET_ID=${i}
15
+ export DIRECTION="enA-koA"
16
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
17
+ export LINE_NO_END=$((DATASET_ID * 2500))
18
+ echo ${LINE_NO_START}
19
+ python fetch_dataset_s2s.py
20
+ done
21
+ for i in $(seq 51 100);
22
+ do
23
+ export N_POOL=15
24
+ export DATASET_ID=${i}
25
+ export DIRECTION="enA-koA"
26
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
27
  export LINE_NO_END=$((DATASET_ID * 2500))
28
  echo ${LINE_NO_START}
29
  python fetch_dataset_s2s.py
30
  done
31
  # tokenize
32
+ for i in $(seq 1 20);
33
  do
34
+ export N_POOL=15
35
  export DATASET_ID=${i}
36
+ export DIRECTION="enA-koA"
37
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
38
+ export LINE_NO_END=$((DATASET_ID * 2500))
39
+ echo ${LINE_NO_START}
40
+ python tokenize_dataset_s2s.py
41
+ done
42
+ for i in $(seq 51 64);
43
+ do
44
+ export N_POOL=15
45
+ export DATASET_ID=${i}
46
+ export DIRECTION="enA-koA"
47
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
48
+ export LINE_NO_END=$((DATASET_ID * 2500))
49
+ echo ${LINE_NO_START}
50
  python tokenize_dataset_s2s.py
51
  done
52
  # speaker embedding
53
+ for i in $(seq 1 20);
54
  do
55
  export DATASET_ID=${i}
56
+ export DIRECTION="enA-koA"
57
+ python attach_speaker_embedding_s2s.py
58
+ done
59
+ for i in $(seq 51 64);
60
+ do
61
+ export DATASET_ID=${i}
62
+ export DIRECTION="enA-koA"
63
  python attach_speaker_embedding_s2s.py
64
  done
65
 
66
+ ######################
67
+ # enA-zhA: 1_289_192 #
68
+ ######################
69
  # test
70
  export DATASET_ID=test
71
+ export DIRECTION="enA-zhA"
72
  export LINE_NO_START=0
73
  export LINE_NO_END=10
74
  python fetch_dataset_s2s.py
75
  # main
76
+ for i in $(seq 1 258);
77
  do
78
  export N_POOL=15
79
  export DATASET_ID=${i}
80
+ export DIRECTION="enA-zhA"
81
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
82
  export LINE_NO_END=$((DATASET_ID * 2500))
83
  echo ${LINE_NO_START}
84
  python fetch_dataset_s2s.py
85
  done
86
  # tokenize
87
+ for i in $(seq 81 100);
88
  do
89
+ export N_POOL=15
90
  export DATASET_ID=${i}
91
+ export DIRECTION="enA-zhA"
92
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
93
+ export LINE_NO_END=$((DATASET_ID * 2500))
94
+ echo ${LINE_NO_START}
95
  python tokenize_dataset_s2s.py
96
  done
97
  # speaker embedding
98
+ for i in $(seq 81 100);
99
  do
100
  export DATASET_ID=${i}
101
+ export DIRECTION="enA-zhA"
102
  python attach_speaker_embedding_s2s.py
103
  done
104
+ ######################
105
+ # enA-frA: 3_054_258 #
106
+ ######################
 
107
  # test
108
  export DATASET_ID=test
109
+ export DIRECTION="enA-frA"
110
  export LINE_NO_START=0
111
  export LINE_NO_END=10
112
  python fetch_dataset_s2s.py
113
  # main
114
+ for i in $(seq 1 300);
115
  do
116
  export N_POOL=15
117
  export DATASET_ID=${i}
118
+ export DIRECTION="enA-frA"
119
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
120
+ export LINE_NO_END=$((DATASET_ID * 2500))
121
+ echo ${LINE_NO_START}
122
+ python fetch_dataset_s2s.py
123
+ done
124
+ for i in $(seq 300 600);
125
+ do
126
+ export N_POOL=15
127
+ export DATASET_ID=${i}
128
+ export DIRECTION="enA-frA"
129
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
130
  export LINE_NO_END=$((DATASET_ID * 2500))
131
  echo ${LINE_NO_START}
132
  python fetch_dataset_s2s.py
133
  done
134
  # tokenize
135
+ for i in $(seq 1 30);
136
  do
137
+ export N_POOL=15
138
  export DATASET_ID=${i}
139
+ export DIRECTION="enA-frA"
140
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
141
+ export LINE_NO_END=$((DATASET_ID * 2500))
142
+ echo ${LINE_NO_START}
143
  python tokenize_dataset_s2s.py
144
  done
145
+ for i in $(seq 300 330);
146
  do
147
+ export N_POOL=15
148
  export DATASET_ID=${i}
149
+ export DIRECTION="enA-frA"
150
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
151
+ export LINE_NO_END=$((DATASET_ID * 2500))
152
+ echo ${LINE_NO_START}
153
  python tokenize_dataset_s2s.py
154
  done
155
  # speaker embedding
156
+ for i in $(seq 1 30);
157
  do
158
  export DATASET_ID=${i}
159
+ export DIRECTION="enA-frA"
160
  python attach_speaker_embedding_s2s.py
161
  done
162
+ # speaker embedding
163
+ for i in $(seq 300 330);
164
+ do
165
+ export DATASET_ID=${i}
166
+ export DIRECTION="enA-frA"
167
+ python attach_speaker_embedding_s2s.py
168
+ done
169
+ ######################
170
+ # enA-deA: 1_965_186 #
171
+ ######################
172
  # test
173
  export DATASET_ID=test
174
+ export DIRECTION="deA-enA"
175
  export LINE_NO_START=0
176
  export LINE_NO_END=10
177
  python fetch_dataset_s2s.py
178
  # main
179
+ for i in $(seq 1 200);
180
  do
181
  export N_POOL=15
182
  export DATASET_ID=${i}
183
+ export DIRECTION="deA-enA"
184
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
185
  export LINE_NO_END=$((DATASET_ID * 2500))
186
  echo ${LINE_NO_START}
187
  python fetch_dataset_s2s.py
188
  done
189
+ for i in $(seq 201 394);
190
  do
191
  export N_POOL=15
192
  export DATASET_ID=${i}
193
+ export DIRECTION="deA-enA"
194
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
195
  export LINE_NO_END=$((DATASET_ID * 2500))
196
  echo ${LINE_NO_START}
197
  python fetch_dataset_s2s.py
198
  done
199
+ # tokenize
200
+ for i in $(seq 1 20);
 
 
 
 
 
 
 
 
 
 
201
  do
202
  export N_POOL=15
203
  export DATASET_ID=${i}
204
+ export DIRECTION="deA-enA"
205
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
206
  export LINE_NO_END=$((DATASET_ID * 2500))
207
  echo ${LINE_NO_START}
208
+ python tokenize_dataset_s2s.py
209
+ done
210
+ for i in $(seq 200 240);
211
+ do
212
+ export N_POOL=15
213
+ export DATASET_ID=${i}
214
+ export DIRECTION="deA-enA"
215
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
216
+ export LINE_NO_END=$((DATASET_ID * 2500))
217
+ echo ${LINE_NO_START}
218
+ python tokenize_dataset_s2s.py
219
+ done
220
+ # speaker embedding
221
+ for i in $(seq 1 20);
222
+ do
223
+ export DATASET_ID=${i}
224
+ export DIRECTION="deA-enA"
225
+ python attach_speaker_embedding_s2s.py
226
+ done
227
+ # speaker embedding
228
+ for i in $(seq 200 240);
229
+ do
230
+ export DATASET_ID=${i}
231
+ export DIRECTION="deA-enA"
232
+ python attach_speaker_embedding_s2s.py
233
  done
 
234
  ######################
235
+ # enA-esA: 2_658_022 #
236
  ######################
237
  # test
238
  export DATASET_ID=test
239
+ export DIRECTION="enA-esA"
240
  export LINE_NO_START=0
241
  export LINE_NO_END=10
242
  python fetch_dataset_s2s.py
243
  # main
244
+ for i in $(seq 1 250);
245
  do
246
  export N_POOL=15
247
  export DATASET_ID=${i}
248
+ export DIRECTION="enA-esA"
249
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
250
  export LINE_NO_END=$((DATASET_ID * 2500))
251
  echo ${LINE_NO_START}
252
  python fetch_dataset_s2s.py
253
  done
254
+ for i in $(seq 251 532);
255
  do
256
  export N_POOL=15
257
  export DATASET_ID=${i}
258
+ export DIRECTION="enA-esA"
259
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
260
  export LINE_NO_END=$((DATASET_ID * 2500))
261
  echo ${LINE_NO_START}
262
  python fetch_dataset_s2s.py
263
  done
264
 
265
+ # tokenize
266
+ for i in $(seq 1 80);
267
+ do
268
+ export N_POOL=15
269
+ export DATASET_ID=${i}
270
+ export DIRECTION="enA-esA"
271
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
272
+ export LINE_NO_END=$((DATASET_ID * 2500))
273
+ echo ${LINE_NO_START}
274
+ python tokenize_dataset_s2s.py
275
+ done
276
+ # speaker embedding
277
+ for i in $(seq 1 80);
278
+ do
279
+ export DATASET_ID=${i}
280
+ export DIRECTION="enA-esA"
281
+ python attach_speaker_embedding_s2s.py
282
+ done
283
+
284
+
285
+ ####################
286
+ # enA-jaA: 718_606 #
287
+ ####################
288
  # test
289
  export DATASET_ID=test
290
+ export DIRECTION="enA-jaA"
291
  export LINE_NO_START=0
292
  export LINE_NO_END=10
293
  python fetch_dataset_s2s.py
294
+ # main
295
+ for i in $(seq 1 144);
296
+ do
297
+ export N_POOL=15
298
+ export DATASET_ID=${i}
299
+ export DIRECTION="enA-jaA"
300
+ export LINE_NO_START=$(((DATASET_ID-1) * 2500))
301
+ export LINE_NO_END=$((DATASET_ID * 2500))
302
+ echo ${LINE_NO_START}
303
+ python fetch_dataset_s2s.py
304
+ done
305
+ # tokenize
306
+ for i in $(seq 1 144);
307
+ do
308
+ export DATASET_ID=${i}
309
+ export DIRECTION="enA-jaA"
310
+ python tokenize_dataset_s2s.py
311
+ done
312
+ # speaker embedding
313
+ for i in $(seq 1 144);
314
+ do
315
+ export DATASET_ID=${i}
316
+ export DIRECTION="enA-jaA"
317
+ python attach_speaker_embedding_s2s.py
318
+ done
319
 
320
+ ####################
321
+ # enA-hiA: 454_942 #
322
+ ####################
323
  # test
324
  export DATASET_ID=test
325
+ export DIRECTION="enA-hiA"
326
  export LINE_NO_START=0
327
  export LINE_NO_END=10
328
  python fetch_dataset_s2s.py
329
  # main
330
+ for i in $(seq 1 91);
331
  do
332
  export N_POOL=15
333
  export DATASET_ID=${i}
334
+ export DIRECTION="enA-hiA"
335
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
336
  export LINE_NO_END=$((DATASET_ID * 2500))
337
  echo ${LINE_NO_START}
338
  python fetch_dataset_s2s.py
339
  done
340
+ # tokenize
341
+ for i in $(seq 1 91);
342
+ do
343
+ export DATASET_ID=${i}
344
+ export DIRECTION="enA-hiA"
345
+ python tokenize_dataset_s2s.py
346
+ done
347
+ # speaker embedding
348
+ for i in $(seq 1 91);
349
+ do
350
+ export DATASET_ID=${i}
351
+ export DIRECTION="enA-hiA"
352
+ python attach_speaker_embedding_s2s.py
353
+ done
354
+ ####################
355
+ # enA-viA: 740_598 #
356
+ ####################
357
+ # test
358
+ export DATASET_ID=test
359
+ export DIRECTION="enA-viA"
360
+ export LINE_NO_START=0
361
+ export LINE_NO_END=10
362
+ python fetch_dataset_s2s.py
363
+ # main
364
+ for i in $(seq 1 149);
365
  do
366
  export N_POOL=15
367
  export DATASET_ID=${i}
368
+ export DIRECTION="enA-viA"
369
  export LINE_NO_START=$(((DATASET_ID-1) * 2500))
370
  export LINE_NO_END=$((DATASET_ID * 2500))
371
  echo ${LINE_NO_START}
372
  python fetch_dataset_s2s.py
373
+ done
374
+ # tokenize
375
+ for i in $(seq 1 149);
376
+ do
377
+ export DATASET_ID=${i}
378
+ export DIRECTION="enA-viA"
379
+ python tokenize_dataset_s2s.py
380
+ done
381
+ # speaker embedding
382
+ for i in $(seq 1 149);
383
+ do
384
+ export DATASET_ID=${i}
385
+ export DIRECTION="enA-viA"
386
+ python attach_speaker_embedding_s2s.py
387
  done
main_s2t.sh CHANGED
@@ -21,6 +21,22 @@ make -j4
21
  alias wet_lines="${PWD}/bin/wet_lines"
22
  cd ../../
23
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
24
  ######################
25
  # enA-jpn: 734_146 #
26
  ######################
@@ -35,7 +51,7 @@ do
35
  done
36
  python format_text.py
37
  # audio
38
- for i in $(seq 61 120);
39
  do
40
  export N_POOL=15
41
  export DATASET_ID=${i}
@@ -46,6 +62,7 @@ do
46
  echo ${LINE_NO_START}
47
  python fetch_dataset_s2t.py
48
  done
 
49
  ########
50
  # NLLB #
51
  ########
 
21
  alias wet_lines="${PWD}/bin/wet_lines"
22
  cd ../../
23
 
24
+
25
+ ######################
26
+ # enA-est: 734_146 #
27
+ ######################
28
+ # text
29
+ export DIRECTION_SPEECH="enA"
30
+ export DIRECTION_TEXT="est"
31
+ export CHUNK_SIZE=20
32
+ python download_s2t_metadata.py
33
+ for i in $(seq 1 ${CHUNK_SIZE});
34
+ do
35
+ cat seamless.dataset.metadata.public.${DIRECTION_SPEECH}-${DIRECTION_TEXT}.withduration.reordered.batch_${i}.tsv | egrep ^crawl-data | tr '\t' ' ' | wet_lines | tee metadata.${DIRECTION_SPEECH}-${DIRECTION_TEXT}.batch_${i}.tsv &
36
+ done
37
+ python format_text.py
38
+
39
+
40
  ######################
41
  # enA-jpn: 734_146 #
42
  ######################
 
51
  done
52
  python format_text.py
53
  # audio
54
+ for i in $(seq 140 294);
55
  do
56
  export N_POOL=15
57
  export DATASET_ID=${i}
 
62
  echo ${LINE_NO_START}
63
  python fetch_dataset_s2t.py
64
  done
65
+
66
  ########
67
  # NLLB #
68
  ########
speaker_embedding_clap.py ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """CLAP embedding.
2
+ - feature dimension: 512
3
+ - source: https://huggingface.co/laion/larger_clap_music_and_speech
4
+ """
5
+ from typing import Optional
6
+
7
+ import torch
8
+ import librosa
9
+ import numpy as np
10
+ from transformers import ClapModel, ClapProcessor
11
+
12
+
13
+ class ClapSE:
14
+ def __init__(self, ckpt: str = "laion/larger_clap_music_and_speech"):
15
+ self.model = ClapModel.from_pretrained(ckpt)
16
+ self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
17
+ self.model.to(self.device)
18
+ self.model.eval()
19
+ self.processor = ClapProcessor.from_pretrained(ckpt)
20
+
21
+ def get_speaker_embedding(self, wav: np.ndarray, sampling_rate: Optional[int] = None) -> np.ndarray:
22
+ if sampling_rate != self.processor.feature_extractor.sampling_rate:
23
+ wav = librosa.resample(wav, orig_sr=sampling_rate, target_sr=self.processor.feature_extractor.sampling_rate)
24
+ inputs = self.processor(
25
+ audios=wav, sampling_rate=self.processor.feature_extractor.sampling_rate, return_tensors="pt"
26
+ )
27
+ with torch.no_grad():
28
+ outputs = self.model.get_audio_features(**{k: v.to(self.device) for k, v in inputs.items()})
29
+ return outputs.cpu().numpy()[0]
30
+
31
+
32
+ class ClapGeneralSE(ClapSE):
33
+
34
+ def __init__(self):
35
+ super().__init__(ckpt="laion/larger_clap_general")