-
Notifications
You must be signed in to change notification settings - Fork 79
/
sptk.py
2293 lines (1661 loc) · 42.5 KB
/
sptk.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
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# coding: utf-8
"""
Library routines
----------------
.. autosummary::
:toctree: generated/
agexp
gexp
glog
mseq
Adaptive cepstrum analysis
--------------------------
.. autosummary::
:toctree: generated/
acep
agcep
amcep
Mel-generalized cepstrum analysis
---------------------------------
.. autosummary::
:toctree: generated/
mcep
gcep
mgcep
uels
fftcep
lpc
MFCC
----
.. autosummary::
:toctree: generated/
mfcc
LPC, LSP and PARCOR conversions
-------------------------------
.. autosummary::
:toctree: generated/
lpc2c
lpc2lsp
lpc2par
par2lpc
lsp2sp
Mel-generalized cepstrum conversions
------------------------------------
.. autosummary::
:toctree: generated/
mc2b
b2mc
c2acr
c2ir
ic2ir
c2ndps
ndps2c
gc2gc
gnorm
ignorm
freqt
mgc2mgc
mgc2sp
mgclsp2sp
F0 analysis
-----------
.. autosummary::
:toctree: generated/
swipe
rapt
Excitation generation
---------------------
.. autosummary::
:toctree: generated/
excite
Window functions
----------------
.. autosummary::
:toctree: generated/
blackman
hamming
hanning
bartlett
trapezoid
rectangular
Waveform generation filters
---------------------------
.. autosummary::
:toctree: generated/
poledf
lmadf
lspdf
ltcdf
glsadf
mlsadf
mglsadf
Utilities for waveform generation filters
-----------------------------------------
.. autosummary::
:toctree: generated/
poledf_delay
lmadf_delay
lspdf_delay
ltcdf_delay
glsadf_delay
mlsadf_delay
mglsadf_delay
"""
import numpy as np
from . import _sptk
### Library routines ###
def agexp(r, x, y):
"""Magnitude squared generalized exponential function
Parameters
----------
r : float
Gamma
x : float
Real part
y : float
Imaginary part
Returns
-------
Value
"""
return _sptk.agexp(r, x, y)
def gexp(r, x):
"""Generalized exponential function
Parameters
----------
r : float
Gamma
x : float
Arg
Returns
-------
Value
"""
return _sptk.gexp(r, x)
def glog(r, x):
"""Generalized logarithmic function
Parameters
----------
r : float
Gamma
x : float
Arg
Returns
-------
Value
"""
return _sptk.glog(r, x)
def mseq():
"""M-sequence
Returns
-------
A sample of m-sequence
"""
return _sptk.mseq()
### Adaptive mel-generalized cepstrum analysis ###
def acep(x, c, lambda_coef=0.98, step=0.1, tau=0.9, pd=4, eps=1.0e-6):
"""Adaptive cepstral analysis
Parameters
----------
x : double
A input sample
c : array, shape(``order + 1``)
Cepstrum. The result is stored in place.
lambda_coef : float, optional
Leakage factor. Default is 0.98.
step : float, optional
Step size. Default is 0.1.
tau : float, optional
Momentum constant. Default is 0.9.
pd : int, optional
Order of pade approximation. Default is 4.
eps : float, optional
Minimum value for epsilon. Default is 1.0e-6.
Returns
-------
prederr : float
Prediction error
Raises
------
ValueError
if invalid order of pade approximation is specified
See Also
--------
pysptk.sptk.uels
pysptk.sptk.gcep
pysptk.sptk.mcep
pysptk.sptk.mgcep
pysptk.sptk.amcep
pysptk.sptk.agcep
pysptk.sptk.lmadf
"""
return _sptk.acep(x, c, lambda_coef, step, tau, pd, eps)
def agcep(x, c, stage=1, lambda_coef=0.98, step=0.1, tau=0.9, eps=1.0e-6):
"""Adaptive generalized cepstral analysis
Parameters
----------
x : float
A input sample
c : array, shape(``order + 1``), optional
Cepstrum. The result is stored in-place.
stage : int, optional
-1 / gamma. Default is 1.
lambda_coef : float, optional
Leakage factor. Default is 0.98.
step : float, optional
Step size. Default is 0.1.
tau : float, optional
Momentum constant. Default is 0.9.
eps : float, optional
Minimum value for epsilon. Default is 1.0e-6.
Returns
-------
prederr : float
Prediction error
Raises
------
ValueError
if invalid number of stage is specified
See Also
--------
pysptk.sptk.acep
pysptk.sptk.amcep
pysptk.sptk.glsadf
"""
return _sptk.agcep(x, c, stage, lambda_coef, step, tau, eps)
def amcep(x, b, alpha=0.35, lambda_coef=0.98, step=0.1, tau=0.9, pd=4, eps=1.0e-6):
"""Adaptive mel-cepstral analysis
Parameters
----------
x : float
A input sample
b : array, shape(``order + 1``), optional
MLSA filter coefficients. The result is stored in-place.
alpha : float, optional
All-pass constant. Default is 0.35.
lambda_coef : float, optional
Leakage factor. Default is 0.98.
step : float, optional
Step size. Default is 0.1.
tau : float, optional
Momentum constant. Default is 0.9.
pd : int, optional
Order of pade approximation. Default is 4.
eps : float, optional
Minimum value for epsilon. Default is 1.0e-6.
Returns
-------
prederr : float
Prediction error
Raises
------
ValueError
if invalid order of pade approximation is specified
See Also
--------
pysptk.sptk.acep
pysptk.sptk.agcep
pysptk.sptk.mc2b
pysptk.sptk.b2mc
pysptk.sptk.mlsadf
"""
return _sptk.amcep(x, b, alpha, lambda_coef, step, tau, pd, eps)
### Mel-generalized cepstrum analysis ###
def mcep(windowed,
order=25, alpha=0.35,
miniter=2,
maxiter=30,
threshold=0.001,
etype=0,
eps=0.0,
min_det=1.0e-6,
itype=0):
"""Mel-cepstrum analysis
Parameters
----------
windowed : array, shape (``frame_len``)
A windowed frame
order : int, optional
Order of mel-cepstrum. Default is 25.
alpha : float, optional
All pass constant. Default is 0.35.
miniter : int, optional
Minimum number of iteration. Default is 2.
maxiter : int, optional
Maximum number of iteration. Default is 30.
threshold : float, optional
Threshold in theq. Default is 0.001.
etype : int, optional
Type of parameter ``eps``
(0) not used
(1) initial value of log-periodogram
(2) floor of periodogram in db
Default is 0.
eps : float, optional
Initial value for log-periodogram or floor of periodogram in db.
Default is 0.0.
min_det : float, optional
Mimimum value of the determinant of normal matrix.
Default is 1.0e-6
itype : float, optional
Input data type
(0) windowed signal
(1) log amplitude in db
(2) log amplitude
(3) amplitude
(4) periodogram
Default is 0.
Returns
-------
mc : array, shape (``order + 1``)
Mel-cepstrum
Raises
------
ValueError
- if invalid ``itype`` is specified
- if invalid ``etype`` is specified
- if nonzero ``eps`` is specified when etype = 0
- if negative ``eps`` is specified
- if negative ``min_det`` is specified
RuntimeError
- if zero(s) are found in periodogram
- if error happened in theq
See Also
--------
pysptk.sptk.uels
pysptk.sptk.gcep
pysptk.sptk.mgcep
pysptk.sptk.mlsadf
"""
return _sptk.mcep(windowed, order, alpha, miniter, maxiter, threshold,
etype, eps, min_det, itype)
def gcep(windowed, order=25, gamma=0.0,
miniter=2,
maxiter=30,
threshold=0.001,
etype=0,
eps=0.0,
min_det=1.0e-6,
itype=0,
norm=False):
"""Generalized-cepstrum analysis
Parameters
----------
windowed : array, shape (``frame_len``)
A windowed frame
order : int, optional
Order of generalized-cepstrum. Default is 25.
gamma : float, optional
Parameter of generalized log function. Default is 0.0.
miniter : int, optional
Minimum number of iteration. Default is 2.
maxiter : int, optional
Maximum number of iteration. Default is 30.
threshold : float, optional
Threshold in theq. Default is 0.001
etype : int, optional
Type of parameter ``eps``
(0) not used
(1) initial value of log-periodogram
(2) floor of periodogram in db
Default is 0.
eps : float, optional
Initial value for log-periodogram or floor of periodogram in db.
Default is 0.0.
min_det : float, optional
Mimimum value of the determinant of normal matrix. Default is 1.0e-6.
itype : float, optional
Input data type
(0) windowed signal
(1) log amplitude in db
(2) log amplitude
(3) amplitude
(4) periodogram
Default is 0.
Returns
-------
gc : array, shape (``order + 1``)
Generalized cepstrum
Raises
------
ValueError
- if invalid ``itype`` is specified
- if invalid ``etype`` is specified
- if nonzero ``eps`` is specified when etype = 0
- if negative ``eps`` is specified
- if negative ``min_det`` is specified
RuntimeError
- if error happened in theq
See Also
--------
pysptk.sptk.uels
pysptk.sptk.mcep
pysptk.sptk.mgcep
pysptk.sptk.glsadf
"""
return _sptk.gcep(windowed, order, gamma, miniter, maxiter, threshold,
etype, eps, min_det, itype)
def mgcep(windowed, order=25, alpha=0.35, gamma=0.0,
num_recursions=None,
miniter=2,
maxiter=30,
threshold=0.001,
etype=0,
eps=0.0,
min_det=1.0e-6,
itype=0,
otype=0):
"""Mel-generalized cepstrum analysis
Parameters
----------
windowed : array, shape (``frame_len``)
A windowed frame
order : int, optional
Order of mel-generalized cepstrum. Default is 25.
alpha : float, optional
All pass constant. Default is 0.35.
gamma : float, optional
Parameter of generalized log function. Default is 0.0.
num_recursions : int, optional
Number of recursions. Default is ``len(windowed) - 1``.
miniter : int, optional
Minimum number of iteration. Default is 2.
maxiter : int, optional
Maximum number of iteration. Default is 30.
threshold : float, optional
Threshold. Default is 0.001.
etype : int, optional
Type of paramter ``e``
(0) not used
(1) initial value of log-periodogram
(2) floor of periodogram in db
Default is 0.
eps : float, optional
Initial value for log-periodogram or floor of periodogram in db.
Default is 0.0.
min_det : float, optional
Mimimum value of the determinant of normal matrix.
Default is 1.0e-6.
itype : float, optional
Input data type
(0) windowed signal
(1) log amplitude in db
(2) log amplitude
(3) amplitude
(4) periodogram
Default is 0.
otype : int, optional
Output data type
(0) mel generalized cepstrum: (c~0...c~m)
(1) MGLSA filter coefficients: b0...bm
(2) K~,c~'1...c~'m
(3) K,b'1...b'm
(4) K~,g*c~'1...g*c~'m
(5) K,g*b'1...g*b'm
Default is 0.
Returns
-------
mgc : array, shape (``order + 1``)
mel-generalized cepstrum
Raises
------
ValueError
- if invalid ``itype`` is specified
- if invalid ``etype`` is specified
- if nonzero ``eps`` is specified when etype = 0
- if negative ``eps`` is specified
- if negative ``min_det`` is specified
- if invalid ``otype`` is specified
RuntimeError
- if error happened in theq
See Also
--------
pysptk.sptk.uels
pysptk.sptk.gcep
pysptk.sptk.mcep
pysptk.sptk.freqt
pysptk.sptk.gc2gc
pysptk.sptk.mgc2mgc
pysptk.sptk.gnorm
pysptk.sptk.mglsadf
"""
return _sptk.mgcep(windowed, order, alpha, gamma, num_recursions, miniter,
maxiter, threshold, etype, eps, min_det, itype, otype)
def uels(windowed, order=25,
miniter=2,
maxiter=30,
threshold=0.001,
etype=0,
eps=0.0,
itype=0):
"""Unbiased estimation of log spectrum
Parameters
----------
windowed : array, shape (``frame_len``)
A windowed frame
order : int, optional
Order of cepstrum. Default is 25.
miniter : int, optional
Minimum number of iteration. Default is 2.
maxiter : int, optional
Maximum number of iteration. Default is 30.
threshold : float, optional
Threshold in theq. Default is 0.001
etype : int, optional
Type of parameter ``eps``
(0) not used
(1) initial value of log-periodogram
(2) floor of periodogram in db
Default is 0.
eps : float, optional
Initial value for log-periodogram or floor of periodogram in db.
Default is 0.0.
itype : float, optional
Input data type
(0) windowed signal
(1) log amplitude in db
(2) log amplitude
(3) amplitude
(4) periodogram
Default is 0.
Returns
-------
c : array, shape (``order + 1``)
cepstrum estimated by uels
Raises
------
ValueError
- if invalid ``itype`` is specified
- if invalid ``etype`` is specified
- if nonzero ``eps`` is specified when etype = 0
- if negative ``eps`` is specified
RuntimeError
- if zero(s) are found in periodogram
See Also
--------
pysptk.sptk.gcep
pysptk.sptk.mcep
pysptk.sptk.mgcep
pysptk.sptk.lmadf
"""
return _sptk.uels(windowed, order, miniter, maxiter, threshold, etype, eps,
itype)
def fftcep(logsp,
order=25,
num_iter=0,
acceleration_factor=0.0):
"""FFT-based cepstrum analysis
Parameters
----------
logsp : array, shape (``frame_len``)
Log power spectrum
order : int, optional
Order of cepstrum. Default is 25.
num_iter : int, optional
Number of iteration. Default is 0.
acceleration_factor : float, optional
Acceleration factor. Default is 0.0.
Returns
-------
c : array, shape (``order + 1``)
Cepstrum
See Also
--------
pysptk.sptk.uels
"""
return _sptk.fftcep(logsp, order, num_iter, acceleration_factor)
def lpc(windowed, order=25, min_det=1.0e-6):
"""Linear prediction analysis
Parameters
----------
windowed : array, shape (``frame_len``)
A windowed frame
order : int, optional
Order of LPC. Default is 25.
min_det : float, optional
Mimimum value of the determinant of normal matrix.
Default is 1.0e-6.
Returns
-------
a : array, shape (``order + 1``)
LPC
Raises
------
ValueError
- if negative ``min_det`` is specified
RuntimeError
- if error happened in levdur
See Also
--------
pysptk.sptk.lpc2par
pysptk.sptk.par2lpc
pysptk.sptk.lpc2c
pysptk.sptk.lpc2lsp
pysptk.sptk.ltcdf
pysptk.sptk.lspdf
"""
return _sptk.lpc(windowed, order, min_det)
### MFCC ###
def mfcc(x, order=14, fs=16000, alpha=0.97, eps=1.0, window_len=None,
frame_len=None, num_filterbanks=20, cepslift=22, use_dft=False,
use_hamming=False, czero=False, power=False):
"""MFCC
Parameters
----------
x : array
A input signal
order : int, optional
Order of MFCC. Default is 14.
fs : int, optional
Sampling frequency. Default is 160000.
alpha : float, optional
Pre-emphasis coefficient. Default is 0.97.
eps : float, optional
Flooring value for calculating ``log(x)`` in filterbank analysis.
Default is 1.0.
window_len : int, optional
Window lenght. Default is ``len(x)``.
frame_len : int, optional
Frame length. Default is ``len(x)``.
num_filterbanks : int, optional
Number of mel-filter banks. Default is 20.
cepslift : int, optional
Liftering coefficient. Default is 22.
use_dft : bool, optional
Use DFT (not FFT) or not. Default is False.
use_hamming : bool, optional
Use hamming window or not. Default is False.
czero : bool, optional
If True, ``mfcc`` returns 0-th coefficient as well. Default is False.
power : bool, optional
If True, ``mfcc`` returns power coefficient as well. Default is False.
Returns
-------
cc : array
MFCC vector, which is ordered as:
mfcc[0], mfcc[1], mfcc[2], ... mfcc[order-1], c0, Power.
Note that c0 and Power are optional.
Shape of ``cc`` is:
- ``order`` by default.
- ``orde + 1`` if ``czero`` or ``power`` is set to True.
- ``order + 2`` if both ``czero`` and ``power`` is set to True.
Raises
------
ValueError
if ``num_filterbanks`` is less than or equal to ``order``
See Also
--------
pysptk.sptk.gcep
pysptk.sptk.mcep
pysptk.sptk.mgcep
"""
return _sptk.mfcc(x, order, fs, alpha, eps, window_len,
frame_len, num_filterbanks, cepslift, use_dft,
use_hamming, czero, power)
### LPC, LSP and PARCOR conversions ###
def lpc2c(lpc, order=None):
"""LPC to cepstrum
Parameters
----------
lpc : array
LPC
order : int, optional
Order of cepstrum. Default is ``len(lpc) - 1``.
Returns
-------
ceps : array, shape (``order + 1``)
cepstrum
See Also
--------
pysptk.sptk.lpc
pysptk.sptk.lspdf
"""
return _sptk.lpc2c(lpc, order)
def lpc2lsp(lpc, numsp=512, maxiter=4, eps=1.0e-6, loggain=False, otype=0,
fs=None):
"""LPC to LSP
Parameters
----------
lpc : array
LPC
numsp : int, optional
Number of unit circle. Default is 512.
maxiter : int, optional
Maximum number of iteration. Default is 4.
eps : float, optional
End condition for iteration. Default is 1.0e-6.
loggain : bool, optional
whether the converted lsp should have loggain or not.
Default is False.
fs : int, optional
Sampling frequency. Default is None and unused.
otype : int, optional
Output format LSP
(0) normalized frequency (0 ~ pi)
(1) normalized frequency (0 ~ 0.5)
(2) frequency (kHz)
(3) frequency (Hz)
Default is 0.
Returns
-------
lsp : array, shape (``order + 1``)
LSP
raises
------
ValueError
if ``fs`` is not specified when otype = 2 or 3.
See Also
--------
pysptk.sptk.lpc
pysptk.sptk.lspdf
"""
return _sptk.lpc2lsp(lpc, numsp, maxiter, eps, loggain, otype, fs)
def lpc2par(lpc):
"""LPC to PARCOR
Parameters
----------
lpc : array
LPC
Returns
-------
par : array, shape (same as ``lpc``)
PARCOR
See Also
--------
pysptk.sptk.lpc
pysptk.sptk.par2lpc
pysptk.sptk.ltcdf
"""
return _sptk.lpc2par(lpc)
def par2lpc(par):
"""PARCOR to LPC
Parameters
----------
par : array
PARCOR
Returns
-------
lpc : array, shape (same as ``par``)
LPC
See Also
--------
pysptk.sptk.lpc
pysptk.sptk.lpc2par
"""
return _sptk.par2lpc(par)