Comments (5)
mravanelli
commented on May 19, 2024
Hi,
we suggest users to compute test alignments only because they might want to
monitor the performance on the test set at some point.
Mirco
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
…On Thu, 7 Mar 2019 at 21:48, JohnJiang ***@***.***> wrote:
*Hi , Mirco. In step 4 of TIMIT tutorial, we should compute the alignments
for test set. Usually, I only calculate MFCC features for test data. . Why
do you do this step?*
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#62>, or mute the
thread
<https://github.com/notifications/unsubscribe-auth/AQGs1q-OFxave8EyPvfODsdn26xOrlxwks5vUc95gaJpZM4bkiPM>
.
from pytorch-kaldi.
Johe-cqu
commented on May 19, 2024
Thank you! @mravanelli
I tried to use mfcc features for test set in this demo.This demo does not seem to support the use of mfcc features.
I think the load_dataset function indata_io.pyneeds to be changed if I want to use mfcc features. But I don't know the format of data_set.
if not fea_only:
lab = { k:v for k,v in read_vec_int_ark('gunzip -c '+lab_folder+'/ali*.gz | '+lab_opts+' '+lab_folder+'/final.mdl ark:- ark:-|',output_folder) if k in fea} # Note that I'm copying only the aligments of the loaded fea
fea = {k: v for k, v in fea.items() if k in lab} # This way I remove all the features without an aligment (see log file in alidir "Did not Succeded")
How should I modify this function ?
John
from pytorch-kaldi.
Johe-cqu
commented on May 19, 2024
>>> read_lab_fea("/home/server/pytorch-kaldi-thchs30/exp/thchs30_GRU_mfcc_high/exp_files/forward_thchs30_test_ep4_ck0.cfg","False",shared_list,"/home/server/pytorch-kaldi-thchs30/test/")
>>> print(shared_list)
[
['D13_919', 'D31_892', 'D12_919', 'D12_880', 'D31_886', 'D12_951', 'D12_981', 'D12_917', 'D12_794', 'D08_794', 'D08_879', 'D12_825', 'D12_987', 'D04_792', 'D08_874', 'D08_915', 'D31_966', 'D12_897', 'D06_938', 'D31_914', 'D06_794', 'D12_986', 'D12_831', 'D06_933', 'D11_880', 'D31_880', 'D32_777', 'D12_884', 'D08_985', 'D12_826', 'D11_797', 'D06_889', 'D12_855', 'D12_837', 'D12_887', 'D06_951', 'D08_864', 'D13_781', 'D13_939', 'D21_968', 'D12_819', 'D12_810', 'D12_827', 'D12_849', 'D06_752', 'D12_926', 'D12_781', 'D12_955', 'D08_986', 'D32_874', 'D12_763', 'D13_773', 'D07_919', 'D12_873', 'D04_968', 'D11_892', 'D13_884', 'D13_931', 'D12_961', 'D12_854', 'D12_816', 'D07_968', 'D08_981', 'D12_838', 'D31_969', 'D11_767', 'D13_825', 'D21_892', 'D06_893', 'D21_879', 'D12_770', 'D31_817', 'D07_751', 'D08_922', 'D08_989', 'D31_855', 'D08_890', 'D12_899', 'D13_922', 'D31_764', 'D31_938', 'D32_884', 'D12_969', 'D07_985', 'D08_928', 'D08_929', 'D13_975', 'D12_820', 'D06_918', 'D07_966', 'D11_919', 'D13_824', 'D31_799', 'D06_907', 'D06_996', 'D31_763', 'D31_921', 'D12_857', 'D06_876', 'D08_846', 'D08_882', 'D08_887', 'D12_752', 'D12_806', 'D13_778', 'D32_771', 'D11_819', 'D12_811', 'D06_783', 'D06_823', 'D06_949', 'D06_967', 'D31_984', 'D08_782', 'D08_808', 'D13_859', 'D13_974', 'D31_926', 'D31_940', 'D04_803', 'D08_920', 'D08_967', 'D11_934', 'D13_842', 'D12_875', 'D12_818', 'D06_826', 'D06_863', 'D06_977', 'D08_771', 'D08_999', 'D13_894', 'D31_760', 'D32_881', 'D12_979', 'D11_813', 'D11_902', 'D06_970', 'D07_955', 'D31_911', 'D11_843', 'D13_760', 'D06_761', 'D07_989', 'D08_854', 'D11_894', 'D13_841', 'D13_888', 'D31_950', 'D04_764', 'D07_783', 'D07_952', 'D08_833', 'D08_837', 'D08_996', 'D12_750', 'D13_871', 'D13_928', 'D31_882', 'D31_987', 'D12_941', 'D12_802', 'D13_977', 'D06_831', 'D07_781', 'D08_930', 'D31_947', 'D12_950', 'D12_957', 'D12_965', 'D04_891', 'D06_763', 'D06_859', 'D06_896', 'D08_800', 'D31_823', 'D31_901', 'D31_907', 'D12_924', 'D11_837', 'D11_838', 'D13_882', 'D21_759', 'D21_981', 'D04_855', 'D08_859', 'D08_867', 'D08_830', 'D11_893', 'D13_857', 'D13_981', 'D31_813', 'D31_965', 'D12_988', 'D12_925', 'D13_769', 'D13_860', 'D04_901', 'D04_966', 'D06_916', 'D08_791', 'D08_865', 'D08_961', 'D31_975', 'D32_781', 'D11_827', 'D13_984', 'D06_791', 'D06_848', 'D07_874', 'D08_907', 'D08_918', 'D31_796', 'D31_824', 'D32_765', 'D12_798', 'D13_891', 'D13_973', 'D21_755', 'D04_895', 'D06_801', 'D07_773', 'D07_948', 'D08_765', 'D08_888', 'D31_854', 'D31_895', 'D32_939', 'D21_884', 'D06_816', 'D06_865', 'D06_965', 'D07_986', 'D08_978', 'D31_784', 'D11_752', 'D04_918', 'D31_821', 'D32_808', 'D13_806', 'D04_989', 'D06_789', 'D07_764', 'D07_922', 'D08_810', 'D13_804', 'D13_903', 'D21_764', 'D06_776', 'D07_755', 'D32_799', 'D32_878', 'D11_836', 'D11_857', 'D21_880', 'D04_929', 'D06_757', 'D07_930', 'D08_828', 'D31_809', 'D31_989', 'D32_787', 'D11_889', 'D21_996', 'D31_828', 'D32_770', 'D32_890', 'D11_952', 'D04_769', 'D04_990', 'D06_811', 'D06_818', 'D06_935', 'D07_767', 'D07_770', 'D07_959', 'D08_842', 'D08_875', 'D31_972', 'D32_776', 'D11_997', 'D13_851', 'D04_993', 'D08_847', 'D31_860', 'D32_966', 'D11_803', 'D21_976', 'D06_758', 'D06_904', 'D07_987', 'D12_906', 'D11_762', 'D11_780', 'D11_887', 'D13_826', 'D06_829', 'D07_894', 'D11_927', 'D11_936', 'D13_865', 'D13_960', 'D21_934', 'D21_967', 'D06_936', 'D07_882', 'D08_903', 'D32_800', 'D32_914', 'D13_776', 'D13_811', 'D04_813', 'D07_920', 'D32_835', 'D32_854', 'D32_882', 'D11_984', 'D04_958', 'D31_808', 'D21_858', 'D21_881', 'D04_896', 'D04_956', 'D04_976', 'D06_960', 'D08_827', 'D08_953', 'D32_871', 'D21_788', 'D04_791', 'D04_842', 'D06_941', 'D31_822', 'D32_856', 'D04_811', 'D07_780', 'D07_876', 'D07_982', 'D08_960', 'D11_961', 'D04_932', 'D07_791', 'D11_758', 'D11_759', 'D11_903', 'D04_787', 'D04_908', 'D06_868', 'D06_962', 'D07_808', 'D08_971', 'D11_948', 'D32_895', 'D11_986', 'D11_916', 'D13_772', 'D04_808', 'D04_979', 'D07_852', 'D13_866', 'D21_930', 'D04_907', 'D04_931', 'D07_834', 'D32_911', 'D11_800', 'D21_770', 'D21_808', 'D04_888', 'D07_849', 'D07_997', 'D08_941', 'D31_754', 'D31_769', 'D21_750', 'D21_753', 'D21_851', 'D07_756', 'D07_820', 'D07_957', 'D32_848', 'D32_927', 'D11_975', 'D21_815', 'D04_822', 'D07_869', 'D07_871', 'D07_902', 'D07_810', 'D31_913', 'D11_896', 'D07_787', 'D07_895', 'D21_893', 'D06_756', 'D11_958', 'D13_786', 'D21_795', 'D21_828', 'D21_989', 'D11_982', 'D32_840', 'D32_906', 'D21_878', 'D04_962', 'D12_964', 'D11_891', 'D13_787', 'D13_816', 'D21_833', 'D21_902', 'D07_883', 'D31_909', 'D32_955', 'D13_906', 'D21_813', 'D04_805', 'D04_829', 'D06_786', 'D07_822', 'D07_867', 'D07_906', 'D07_943', 'D07_965', 'D13_798', 'D21_961', 'D07_910', 'D11_957', 'D11_760', 'D21_888', 'D21_891', 'D21_924', 'D04_864', 'D07_848', 'D11_911', 'D21_908', 'D21_925', 'D32_921', 'D13_965', 'D21_820', 'D21_912', 'D32_844', 'D21_832', 'D04_775', 'D32_918', 'D21_862', 'D07_904', 'D32_814', 'D32_898', 'D21_854', 'D04_814', 'D21_857', 'D21_982', 'D21_990', 'D21_836', 'D31_941', 'D21_885', 'D04_772', 'D07_758', 'D07_774', 'D08_909', 'D11_941', 'D21_957', 'D21_900', 'D07_960', 'D13_991', 'D21_756', 'D21_972', 'D07_850', 'D07_868', 'D21_920', 'D21_973', 'D21_847', 'D06_998', 'D21_853', 'D04_885', 'D21_903', 'D04_903', 'D21_818', 'D07_786', 'D21_935', 'D21_950', 'D04_860', 'D04_925', 'D11_885', 'D21_936', 'D21_937', 'D21_754', 'D32_982', 'D11_980', 'D32_997', 'D08_766', 'D21_962', 'D04_923', 'D07_980', 'D32_923', 'D21_906', 'D32_991', 'D21_923'], 数据名称data_name
array([ 489, 1012, 1536, 2091, 2664, 3250, 3850, 4452,
5058, 5675, 6292, 6914, 7540, 8176, 8812, 9448,
10084, 10723, 11365, 12007, 12655, 13305, 13963, 14624,
15285, 15946, 16607, 17269, 17936, 18603, 19271, 19944,
20618, 21294, 21973, 22659, 23345, 24031, 24717, 25403,
26091, 26780, 27471, 28162, 28854, 29552, 30251, 30954,
31658, 32362, 33067, 33772, 34483, 35198, 35915, 36632,
37349, 38066, 38787, 39512, 40240, 40969, 41698, 42427,
43156, 43886, 44616, 45346, 46082, 46818, 47560, 48302,
49050, 49798, 50546, 51294, 52048, 52802, 53556, 54310,
55064, 55818, 56575, 57336, 58097, 58858, 59619, 60383,
61150, 61917, 62684, 63451, 64218, 64991, 65764, 66537,
67310, 68084, 68863, 69642, 70421, 71200, 71979, 72758,
73537, 74316, 75101, 75886, 76672, 77458, 78244, 79030,
79816, 80608, 81400, 82192, 82984, 83776, 84568, 85366,
86164, 86962, 87760, 88558, 89359, 90162, 90966, 91770,
92574, 93378, 94182, 94986, 95790, 96594, 97403, 98213,
99023, 99834, 100645, 101456, 102272, 103088, 103911, 104734,
105557, 106380, 107203, 108026, 108849, 109678, 110507, 111336,
112165, 112994, 113823, 114652, 115481, 116310, 117139, 117968,
118800, 119635, 120470, 121306, 122142, 122978, 123814, 124653,
125492, 126331, 127173, 128015, 128857, 129699, 130541, 131383,
132225, 133067, 133911, 134758, 135605, 136452, 137299, 138146,
138994, 139842, 140690, 141544, 142398, 143252, 144106, 144960,
145814, 146669, 147525, 148385, 149245, 150106, 150967, 151828,
152689, 153550, 154411, 155272, 156133, 156999, 157865, 158732,
159599, 160466, 161333, 162200, 163067, 163934, 164801, 165673,
166545, 167417, 168289, 169162, 170035, 170908, 171781, 172654,
173527, 174400, 175273, 176146, 177024, 177903, 178782, 179661,
180540, 181419, 182298, 183183, 184069, 184955, 185841, 186732,
187624, 188516, 189408, 190300, 191192, 192089, 192986, 193883,
194781, 195679, 196577, 197475, 198378, 199281, 200184, 201088,
201992, 202896, 203800, 204704, 205608, 206512, 207421, 208330,
209241, 210152, 211063, 211977, 212894, 213811, 214728, 215645,
216562, 217479, 218396, 219313, 220230, 221147, 222064, 222981,
223900, 224822, 225745, 226668, 227591, 228514, 229442, 230370,
231299, 232228, 233157, 234089, 235023, 235957, 236891, 237825,
238761, 239697, 240637, 241577, 242517, 243457, 244397, 245337,
246279, 247221, 248163, 249105, 250047, 250994, 251941, 252889,
253837, 254785, 255733, 256681, 257633, 258587, 259541, 260500,
261459, 262420, 263381, 264342, 265303, 266264, 267225, 268186,
269151, 270118, 271085, 272052, 273019, 273986, 274959, 275932,
276905, 277878, 278851, 279828, 280807, 281786, 282770, 283754,
284738, 285724, 286710, 287696, 288682, 289668, 290654, 291640,
292626, 293615, 294605, 295595, 296587, 297579, 298571, 299567,
300563, 301561, 302559, 303557, 304555, 305557, 306559, 307561,
308565, 309569, 310573, 311577, 312581, 313585, 314594, 315603,
316612, 317623, 318634, 319645, 320656, 321667, 322680, 323695,
324712, 325729, 326746, 327763, 328786, 329809, 330836, 331865,
332894, 333927, 334963, 336001, 337041, 338081, 339121, 340161,
341203, 342245, 343287, 344333, 345381, 346432, 347484, 348536,
349588, 350640, 351692, 352746, 353800, 354854, 355912, 356970,
358031, 359092, 360153, 361214, 362275, 363336, 364397, 365458,
366522, 367586, 368653, 369721, 370792, 371863, 372934, 374005,
375078, 376151, 377228, 378305, 379382, 380461, 381544, 382627,
383710, 384796, 385885, 386977, 388069, 389164, 390268, 391372,
392476, 393584, 394695, 395809, 396923, 398037, 399157, 400280,
401406, 402535, 403664, 404793, 405922, 407055, 408188, 409327,
410469, 411614, 412759, 413904, 415052, 416200, 417351, 418502,
419659, 420820, 421990, 423163, 424339, 425525, 426713, 427905,
429100, 430295, 431493, 432691, 433904, 435117, 436336, 437574,
438816, 440076, 441337, 442610, 443891, 445195, 446512, 447841,
449197, 450664, 452145]),
{'mfcc': ['mfcc', 'exp/thchs30_GRU_mfcc_high/exp_files/forward_thchs30_test_ep4_ck0_mfcc.lst', 'apply-cmvn --utt2spk=ark:/home/server/kaldi/egs/thchs30/s5/data/mfcc/test/utt2spk ark:/home/server/kaldi/egs/thchs30/s5/mfcc/test/cmvn_test.ark ark:- ark:- | add-deltas --delta-order=2 ark:- ark:- |', '5', '5', 0, 429, 429]},
{'lab_name': 'none', 'lab_cd': ['lab_cd', '/home/server/kaldi/egs/thchs30/s5/exp/tri4b_ali_test_phone', 'ali-to-pdf']},
{'GRU_layers': ['architecture1', 'GRU_layers', 1], 'MLP_layers': ['architecture2', 'MLP_layers', 0]},
array([[-1.33263227, -0.79611357, -0.11324069, ..., -0.59629109,
0. , 0. ],
[-1.31921462, -0.93640346, -0.05924353, ..., 0.08606095,
0. , 0. ],
[-1.23870904, -0.90133102, 0.00375313, ..., 0.84479017,
0. , 0. ],
...,
[-0.89798079, 0.31299214, -0.83178704, ..., -0.5393988 ,
0. , 0. ],
[-0.78412175, 0.2488659 , -0.86888197, ..., -1.24421733,
0. , 0. ],
[-0.11723276, -0.99091591, -1.46240041, ..., -0.6490067 ,
0. , 0. ]])
]
>>> data_name=shared_list[0]
>>> data_end_index=shared_list[1]
>>> fea_dict=shared_list[2]
>>> lab_dict=shared_list[3]
>>> arch_dict=shared_list[4]
>>> data_set=shared_list[5]
from pytorch-kaldi.
TParcollet
commented on May 19, 2024
Hi, could you please paste the error that you are seeing ? MFCC should work.
from pytorch-kaldi.
Johe-cqu
commented on May 19, 2024
Sorry, I found some mistakes in my code, I have solved this problem.
from pytorch-kaldi.
Related Issues (20)
- How to setup parameters in "cfg/TIMIT_baselines/TIMIT_liGRU_fmllr.cfg"? HOT 1
- Do bidirectional layers share the input-to-hidden weights? HOT 2
- Can we resume training from the epoch we got interruption HOT 4
- input shape of nns HOT 3
- Question about the Dimension of wx.0.weight in my mlp model HOT 1
- The loss curve of train and dev is reasonable but why the Test Error keeps 53% or so? HOT 8
- Support for torch.nn.Transformer Class? HOT 1
- KaldiFatalError during decoding phase
- No WER stdout when decoding
- Does pytorch-kaldi support chain model training? HOT 1
- Word transcription of TIMIT dataset HOT 1
- No Decoding Output HOT 20
- How to train/decode on reverberant speech? HOT 1
- x-vector DNN model
- Unable to run forwarding step on test set
- Before switch to SpeechBrain, how to use trained model in pytorch
- Use final_architecture1.pkl for live test HOT 4
- err_te is 1
- using different features instead of FMLLR
- res.res
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch-kaldi.