-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
1060 lines (950 loc) · 49.1 KB
/
index.html
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
---
# Feel free to add content and custom Front Matter to this file.
# To modify the layout, see https://jekyllrb.com/docs/themes/#overriding-theme-defaults
layout: default
---
<style>
.video-container {
position: relative;
padding-bottom: 56.25%; /* 16:9 aspect ratio */
height: 0;
overflow: hidden;
max-width: 100%;
background: #000;
}
.video-container iframe {
position: absolute;
top: 0;
left: 0;
width: 100%;
height: 100%;
}
audio {
width: 100%;
}
table {
width: 100%;
border-collapse: collapse;
}
.table-container {
width: 100%;
max-width: 100%;
overflow-x: auto;
-webkit-overflow-scrolling: touch; /* For smooth scrolling on mobile devices */
}
html, body {
margin: 5px; /* Get rid of default margins */
padding: 5px; /* Get rid of default padding */
overflow-x: hidden; /* Prevent horizontal overflow */
}
th, td {
padding: 4px;
text-align: left;
border: 1px solid #ddd;
font-size: 1em; /* Base font size */
white-space: nowrap;
}
p {
text-align: justify;
hyphens: auto;
}
div {
margin: 0;
padding: 0;
box-sizing: border-box;
max-width: 100%;
}
/* Define a class for left-aligned text */
.left-align {
text-align: left;
hyphens: none;
}
.container {
max-width: 100%;
background-color: #f4f4f4;
align-items: left;
overflow-x: auto; /* Handle overflow */
padding: 0;
margin: 0 auto;
position: relative;
}
.inner-container {
width: calc(100% - 30px);
background-color: #f4f4f4;
align-items: left;
padding: 0;
margin: 0 auto;
overflow: hidden;
position: relative; /* Relative positioning for absolute child positioning */
}
pre {
width: calc(100% - 30px);
background-color: #f4f4f4;
border: 0;
padding: 0px;
overflow: auto;
border-radius: 0px;
margin: 0;
}
code {
font-family: monospace;
padding: 0px;
margin: 0;
}
.highlight {
cursor: pointer;
padding: 4px 4px;
background-color: #f4f4f4;
border: 0;
color: white;
border-radius: 4px;
display: flex;
align-items: center;
max-width: 100%;
position: absolute;
top: 0; /* Position the button at the top */
right: 0; /* Position the button at the right edge */
margin: 5px; /* Small margin for aesthetics */
}
.highlight svg {
fill: #4e4e4e;
margin-right: 5px;
}
.highlight:hover {
background-color: #d3d3d3;
}
</style>
<h1>
EARS: An Anechoic Fullband Speech Dataset Benchmarked for Speech Enhancement and Dereverberation
</h1>
<p class="left-align">
Julius Richter, Yi-Chiao Wu, Steven Krenn, Simon Welker, Bunlong Lay, Shinji Watanabe, Alexander Richard, Timo
Gerkmann
</p>
<h2>
Abstract
</h2>
<p>
We release the EARS (<b>E</b>xpressive <b>A</b>nechoic <b>R</b>ecordings of <b>S</b>peech) dataset, a high-quality
speech dataset comprising 107 speakers from diverse backgrounds, totalling in more than 100 hours of clean, anechoic
speech data. The dataset covers a large range of different speaking styles, including emotional speech, different
reading styles, non-verbal sounds, and conversational freeform speech. We benchmark various methods for speech
enhancement and dereverberation on the dataset and evaluate their performance through a set of instrumental metrics.
In addition, we conduct a listening test with 20 participants for the speech enhancement task, where a generative
method is preferred. We introduce a blind test set that allows for automatic online evaluation of uploaded data.
Dataset download links and automatic evaluation server can be found online.
</p>
<h2>
EARS Dataset
</h2>
<p>
The EARS dataset is characterized by its scale, diversity, and high recording quality. In Table 1, we list
characteristics of the EARS dataset in comparison to other speech datasets.
</p>
<div class="table-container">
<table>
<caption style="caption-side: bottom; text-align: left; padding: 8px; font-style: italic;">
<strong>Table 1: Speech datasets.</strong> In contrast to existing datasets, the EARS dataset is of higher
recording quality, large, and more diverse. <sup>†</sup>contains files with limited bandwidth.
</caption>
<thead>
<tr style="border-top: 2px solid black; border-bottom: 2px solid black;">
<th> </th>
<th>hours</th>
<th>speakers</th>
<th>sample rate</th>
</tr>
</thead>
<tbody>
<tr>
<td>DNS (LibriVox) <span class="reference" data-ref="librivox"></span></td>
<td>556</td>
<td>1948</td>
<td>48 kHz<sup>†</sup></td>
</tr>
<tr>
<td>LibriSpeech <span class="reference" data-ref="librispeech"></span></td>
<td>982</td>
<td>2484</td>
<td>16 kHz</td>
</tr>
<tr>
<td>LJSpeech <span class="reference" data-ref="ljspeech"></span></td>
<td>24</td>
<td>1</td>
<td>22.05 kHz</td>
</tr>
<tr>
<td>TIMIT <span class="reference" data-ref="timit"></span></td>
<td>5</td>
<td>632</td>
<td>16 kHz</td>
</tr>
<tr>
<td>VCTK <span class="reference" data-ref="vctk"></span></td>
<td>44</td>
<td>110</td>
<td>48 kHz</td>
</tr>
<tr>
<td>WSJ0 <span class="reference" data-ref="wsj0"></span></td>
<td>29</td>
<td>119</td>
<td>16 kHz</td>
</tr>
<tr style="border-bottom: 2px solid black;">
<td><b>EARS (ours)</b></td>
<td><b>100</b></td>
<td><b>107</b></td>
<td><b>48 kHz</b></td>
</tr>
</tbody>
</table>
</div>
<p>
EARS contains 100 h of anechoic speech recordings at 48 kHz from over 100 English speakers with high demographic
diversity. The dataset spans the full range of human speech, including reading tasks in seven different reading
styles, emotional reading and freeform speech in 22 different emotions, conversational speech, and non-verbal sounds
like laughter or coughing. Reading tasks feature seven styles (regular, loud, whisper, fast, slow, high pitch, and
low pitch). Additionally, the dataset features unconstrained freeform speech and speech in 22 different emotional
styles. We provide transcriptions of the reading portion and meta-data of the speakers (gender, age, race, first
language).
</p>
<h3>
Audio Examples
</h3>
<p>
Here we present a few audio examples from the EARS dataset.
</p>
<p>
p002/emo_adoration_sentences.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p002_emo_adoration_sentences.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p008/emo_contentment_sentences.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p008_emo_contentment_sentences.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p010/emo_cuteness_sentences.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p010_emo_cuteness_sentences.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p011/emo_anger_sentences.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p011_emo_anger_sentences.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p012/rainbow_05_whisper.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p012_rainbow_05_whisper.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p014/rainbow_04_loud.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p014_rainbow_04_loud.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p016/rainbow_03_regular.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p016_rainbow_03_regular.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p017/rainbow_08_fast.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p017_rainbow_08_fast.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p018/vegetative_eating.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p018_vegetative_eating.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p019/vegetative_yawning.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p019_vegetative_yawning.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
<br>
p020/freeform_speech_01.wav<br>
<audio controls>
<source src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears/p020_freeform_speech_01.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
</p>
<br>
<h2>
Benchmarks
</h2>
<p>
The EARS dataset enables various speech processing tasks to be evaluated in a controlled and comparable way. Here, we
present benchmarks for speech enhancement and dereverberation tasks.
</p>
<br>
<h3>
EARS-WHAM
</h3>
<p>
For the task of speech enhancement, we construct the EARS-WHAM dataset, which mixes speech from the EARS dataset
with real noise recordings from the WHAM! dataset <span class="reference" data-ref="wham"></span>. More details can
be found in the <a href="https://arxiv.org/abs/2406.06185" target="_blank">paper</a>.
</p>
<h4>Results</h4>
<div class="table-container">
<table style="width:100%; border-collapse: collapse;">
<caption style="caption-side: bottom; text-align: left; padding: 8px; font-style: italic;">
<strong>Table 2: Results on EARS-WHAM.</strong> Values indicate the mean of the metrics over the test set.
The best results are highlighted in bold.
</caption>
<thead>
<tr style="border-top: 2px solid black; border-bottom: 2px solid black;">
<th></th>
<th style="text-align: right;">POLQA</th>
<th style="text-align: right;">SI-SDR</th>
<th style="text-align: right;">PESQ</th>
<th style="text-align: right;">ESTOI</th>
<th style="text-align: right;">DNSMOS</th>
</tr>
</thead>
<tbody>
<tr>
<td>Noisy</td>
<td style="text-align: right;">1.71</td>
<td style="text-align: right;">5.98</td>
<td style="text-align: right;">1.24</td>
<td style="text-align: right;">0.49</td>
<td style="text-align: right;">2.74</td>
</tr>
<tr>
<td>Conv-TasNet <span class="reference" data-ref="convtasnet"></span></td>
<td style="text-align: right;">2.73</td>
<td style="text-align: right;"><strong>16.93</strong></td>
<td style="text-align: right;">2.31</td>
<td style="text-align: right;">0.70</td>
<td style="text-align: right;">3.47</td>
</tr>
<tr>
<td>CDiffuSE <span class="reference" data-ref="cdiffuse"></span></td>
<td style="text-align: right;">1.81</td>
<td style="text-align: right;">8.35</td>
<td style="text-align: right;">1.60</td>
<td style="text-align: right;">0.53</td>
<td style="text-align: right;">2.87</td>
</tr>
<tr>
<td>Demucs <span class="reference" data-ref="demucs"></span></td>
<td style="text-align: right;">2.97</td>
<td style="text-align: right;">16.92</td>
<td style="text-align: right;">2.37</td>
<td style="text-align: right;">0.71</td>
<td style="text-align: right;">3.66</td>
</tr>
<tr style="border-bottom: 2px solid black;">
<td>SGMSE+ <span class="reference" data-ref="sgmse"></span></td>
<td style="text-align: right;"><strong>3.40</strong></td>
<td style="text-align: right;">16.78</td>
<td style="text-align: right;"><strong>2.50</strong></td>
<td style="text-align: right;"><strong>0.73</strong></td>
<td style="text-align: right;"><strong>3.88</strong></td>
</tr>
</tbody>
</table>
</div>
<h4>
Audio Examples
</h4>
<p>
Here we present audio examples for the speech enhancement task. Below we show the noisy input, processed files for
Conv-TasNet <span class="reference" data-ref="convtasnet"></span>,
CDiffuSE <span class="reference" data-ref="cdiffuse"></span>,
Demucs <span class="reference" data-ref="demucs"></span>,
SGMSE+ <span class="reference" data-ref="sgmse"></span>,
and the clean ground truth.
</p>
<p>Select an audio file:
<select id="audioSelect" onchange="playAudio()">
<option value="p102/00252_4.9dB">p102/00252_4.9dB</option>
<option value="p103/00464_11.8dB">p103/00464_11.8dB</option>
<option value="p104/00814_11.3dB">p104/00814_11.3dB</option>
<option value="p105/00735_0.4dB">p105/00735_0.4dB</option>
<option value="p106/00049_5.4dB">p106/00049_5.4dB</option>
<option value="p107/00519_8.1dB">p107/00519_8.1dB</option>
<option value="p102/00404_4.0dB">p102/00404_4.0dB</option>
<option value="p103/00642_9.7dB">p103/00642_9.7dB</option>
<option value="p104/00014_6.9dB">p104/00014_6.9dB</option>
<option value="p105/00246_3.8dB">p105/00246_3.8dB</option>
<option value="p106/00060_0.7dB">p106/00060_0.7dB</option>
<option value="p107/00159_-0.8dB">p107/00159_-0.8dB</option>
<option value="p102/00189_2.4dB">p102/00189_2.4dB</option>
<option value="p103/00483_1.5dB">p103/00483_1.5dB</option>
<option value="p104/00357_0.5dB">p104/00357_0.5dB</option>
<option value="p105/00574_11.4dB">p105/00574_11.4dB</option>
<option value="p106/00291_5.8dB">p106/00291_5.8dB</option>
<option value="p107/00142_6.0dB">p107/00142_6.0dB</option>
<option value="p102/00327_1.9dB">p102/00327_1.9dB</option>
<option value="p103/00051_11.0dB">p103/00051_11.0dB</option>
<option value="p104/00507_0.6dB">p104/00507_0.6dB</option>
<option value="p105/00655_14.3dB">p105/00655_14.3dB</option>
<option value="p106/00852_14.9dB">p106/00852_14.9dB</option>
<option value="p107/00575_15.6dB">p107/00575_15.6dB</option>
<option value="p102/00785_6.2dB">p102/00785_6.2dB</option>
<option value="p103/00595_5.8dB">p103/00595_5.8dB</option>
<option value="p104/00662_3.4dB">p104/00662_3.4dB</option>
<option value="p105/00148_11.6dB">p105/00148_11.6dB</option>
<option value="p106/00039_3.5dB">p106/00039_3.5dB</option>
<option value="p107/00173_5.9dB">p107/00173_5.9dB</option>
</select>
</p>
<p>
Noisy: <br>
<audio id="audioPlayerNoisy" controls>
<source id="audioSourceNoisy" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/noisy/p102/00252_4.9dB.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
Conv-TasNet <span class="reference" data-ref="convtasnet"></span>:<br>
<audio id="audioPlayerConvTasNet" controls>
<source id="audioSourceConvTasNet" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/convtasnet/p102/00252_4.9dB.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
CDiffuSE <span class="reference" data-ref="cdiffuse"></span>:<br>
<audio id="audioPlayerCDiffuSE" controls>
<source id="audioSourceCDiffuSE" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/cdiffuse/p102/00252_4.9dB.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
Demucs <span class="reference" data-ref="demucs"></span>:<br>
<audio id="audioPlayerDemucs" controls>
<source id="audioSourceDemucs" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/demucs/p102/00252_4.9dB.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
SGMSE+ <span class="reference" data-ref="sgmse"></span>:<br>
<audio id="audioPlayerSGMSE" controls>
<source id="audioSourceSGMSE" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/sgmse/p102/00252_4.9dB.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
Clean: <br>
<audio id="audioPlayerClean" controls>
<source id="audioSourceClean" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/clean/p102/00252_4.9dB.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
</p>
<script>
function playAudio() {
var audioPlayerNoisy = document.getElementById('audioPlayerNoisy');
var audioPlayerConvTasNet = document.getElementById('audioPlayerConvTasNet');
var audioPlayerCDiffuSE = document.getElementById('audioPlayerCDiffuSE');
var audioPlayerDemucs = document.getElementById('audioPlayerDemucs');
var audioPlayerSGMSE = document.getElementById('audioPlayerSGMSE');
var audioPlayerClean = document.getElementById('audioPlayerClean');
var audioSourceNoisy = document.getElementById('audioSourceNoisy');
var audioSourceConvTasNet = document.getElementById('audioSourceConvTasNet');
var audioSourceCDiffuSE = document.getElementById('audioSourceCDiffuSE');
var audioSourceDemucs = document.getElementById('audioSourceDemucs');
var audioSourceSGMSE = document.getElementById('audioSourceSGMSE');
var audioSourceClean = document.getElementById('audioSourceClean');
var selectedAudio = document.getElementById('audioSelect').value;
if (selectedAudio) {
// Set the source of the source elements, not the audio players
audioSourceNoisy.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/noisy/" + selectedAudio + ".wav";
audioPlayerNoisy.load(); // Load the selected audio file using the audio player
audioSourceConvTasNet.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/convtasnet/" + selectedAudio + ".wav";
audioPlayerConvTasNet.load(); // Load the selected audio file using the audio player
audioSourceCDiffuSE.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/cdiffuse/" + selectedAudio + ".wav";
audioPlayerCDiffuSE.load(); // Load the selected audio file using the audio player
audioSourceDemucs.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/demucs/" + selectedAudio + ".wav";
audioPlayerDemucs.load(); // Load the selected audio file using the audio player
audioSourceSGMSE.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/sgmse/" + selectedAudio + ".wav";
audioPlayerSGMSE.load(); // Load the selected audio file using the audio player
audioSourceClean.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-wham/clean/" + selectedAudio + ".wav";
audioPlayerClean.load(); // Load the selected audio file using the audio player
// This will ensure that the audio players are updating their children source elements
// and reloading them
}
}
</script>
<br>
<h3>
Blind test set
</h3>
<p>
We create a blind test set for which we only publish the noisy audio files but not the clean ground truth. It
contains 743 files (2 h) from six speakers (3 male, 3 female) that are not part of the EARS dataset and noise
especially recorded for this test set.
</p>
<h4>Results</h4>
<div class="table-container">
<table>
<caption style="caption-side: bottom; text-align: left; padding: 8px; font-style: italic;">
<strong>Table 3: Results for the blind test.</strong> Values indicate the mean of the metrics over the test
set. The best results are highlighted in bold.
</caption>
<thead>
<tr style="border-top: 2px solid black; border-bottom: 2px solid black;">
<th></th>
<th style="text-align: right;">POLQA</th>
<th style="text-align: right;">SI-SDR</th>
<th style="text-align: right;">PESQ</th>
<th style="text-align: right;">ESTOI</th>
<th style="text-align: right;">DNSMOS</th>
</tr>
</thead>
<tbody>
<tr>
<td>Noisy</td>
<td style="text-align: right;">1.81</td>
<td style="text-align: right;">6.48</td>
<td style="text-align: right;">1.28</td>
<td style="text-align: right;">0.57</td>
<td style="text-align: right;">2.79</td>
</tr>
<tr>
<td>Conv-TasNet <span class="reference" data-ref="convtasnet"></span></td>
<td style="text-align: right;">2.68</td>
<td style="text-align: right;">16.56</td>
<td style="text-align: right;">2.41</td>
<td style="text-align: right;">0.75</td>
<td style="text-align: right;">3.43</td>
</tr>
<tr>
<td>CDiffuSE <span class="reference" data-ref="cdiffuse"></span></td>
<td style="text-align: right;">1.93</td>
<td style="text-align: right;">8.22</td>
<td style="text-align: right;">1.64</td>
<td style="text-align: right;">0.59</td>
<td style="text-align: right;">2.92</td>
</tr>
<tr>
<td>Demucs <span class="reference" data-ref="demucs"></span></td>
<td style="text-align: right;">3.03</td>
<td style="text-align: right;"><strong>16.81</strong></td>
<td style="text-align: right;">2.50</td>
<td style="text-align: right;">0.76</td>
<td style="text-align: right;">3.62</td>
</tr>
<tr style="border-bottom: 2px solid black;">
<td>SGMSE+ <span class="reference" data-ref="sgmse"></span></td>
<td style="text-align: right;"><strong>3.35</strong></td>
<td style="text-align: right;">16.43</td>
<td style="text-align: right;"><strong>2.59</strong></td>
<td style="text-align: right;"><strong>0.78</strong></td>
<td style="text-align: right;"><strong>3.79</strong></td>
</tr>
</tbody>
</table>
</div>
<h4>
Audio Examples
</h4>
<p>
Here we present audio examples for the blind test set. Below we show the noisy input, processed files for
Conv-TasNet <span class="reference" data-ref="convtasnet"></span>,
CDiffuSE <span class="reference" data-ref="cdiffuse"></span>,
Demucs <span class="reference" data-ref="demucs"></span>,
and SGMSE+ <span class="reference" data-ref="sgmse"></span>.
</p>
<p>Select an audio file:
<select id="audioSelectBlind" onchange="playAudioBlind()">
<option value="00620">00620</option>
<option value="00377">00377</option>
<option value="00406">00406</option>
<option value="00267">00267</option>
<option value="00294">00294</option>
<option value="00252">00252</option>
<option value="00065">00065</option>
<option value="00399">00399</option>
<option value="00000">00000</option>
<option value="00474">00474</option>
<option value="00488">00488</option>
<option value="00412">00412</option>
<option value="00226">00226</option>
<option value="00520">00520</option>
<option value="00652">00652</option>
<option value="00470">00470</option>
<option value="00561">00561</option>
<option value="00371">00371</option>
<option value="00388">00388</option>
<option value="00178">00178</option>
<option value="00508">00508</option>
<option value="00711">00711</option>
<option value="00202">00202</option>
<option value="00351">00351</option>
<option value="00437">00437</option>
<option value="00303">00303</option>
<option value="00143">00143</option>
<option value="00262">00262</option>
<option value="00282">00282</option>
<option value="00613">00613</option>
</select>
</p>
<p>
Noisy: <br>
<audio id="audioPlayerBlindNoisy" controls>
<source id="audioSourceBlindNoisy" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/noisy/00620.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
Conv-TasNet <span class="reference" data-ref="convtasnet"></span>:<br>
<audio id="audioPlayerBlindConvTasNet" controls>
<source id="audioSourceBlindConvTasNet" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/convtasnet/00620.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
CDiffuSE <span class="reference" data-ref="cdiffuse"></span>:<br>
<audio id="audioPlayerBlindCDiffuSE" controls>
<source id="audioSourceBlindCDiffuSE" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/cdiffuse/00620.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
Demucs <span class="reference" data-ref="demucs"></span>:<br>
<audio id="audioPlayerBlindDemucs" controls>
<source id="audioSourceBlindDemucs" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/demucs/00620.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
SGMSE+ <span class="reference" data-ref="sgmse"></span>:<br>
<audio id="audioPlayerBlindSGMSE" controls>
<source id="audioSourceBlindSGMSE" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/sgmse/00620.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
</p>
<script>
function playAudioBlind() {
var audioPlayerBlindNoisy = document.getElementById('audioPlayerBlindNoisy');
var audioPlayerBlindConvTasNet = document.getElementById('audioPlayerBlindConvTasNet');
var audioPlayerBlindCDiffuSE = document.getElementById('audioPlayerBlindCDiffuSE');
var audioPlayerBlindDemucs = document.getElementById('audioPlayerBlindDemucs');
var audioPlayerBlindSGMSE = document.getElementById('audioPlayerBlindSGMSE');
var audioSourceBlindNoisy = document.getElementById('audioSourceBlindNoisy');
var audioSourceBlindConvTasNet = document.getElementById('audioSourceBlindConvTasNet');
var audioSourceBlindCDiffuSE = document.getElementById('audioSourceBlindCDiffuSE');
var audioSourceBlindDemucs = document.getElementById('audioSourceBlindDemucs');
var audioSourceBlindSGMSE = document.getElementById('audioSourceBlindSGMSE');
var selectedAudioBlind = document.getElementById('audioSelectBlind').value;
if (selectedAudioBlind) {
// Set the source of the source elements, not the audio players
audioSourceBlindNoisy.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/noisy/" + selectedAudioBlind + ".wav";
audioPlayerBlindNoisy.load(); // Load the selected audio file using the audio player
audioSourceBlindConvTasNet.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/convtasnet/" + selectedAudioBlind + ".wav";
audioPlayerBlindConvTasNet.load(); // Load the selected audio file using the audio player
audioSourceBlindCDiffuSE.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/cdiffuse/" + selectedAudioBlind + ".wav";
audioPlayerBlindCDiffuSE.load(); // Load the selected audio file using the audio player
audioSourceBlindDemucs.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/demucs/" + selectedAudioBlind + ".wav";
audioPlayerBlindDemucs.load(); // Load the selected audio file using the audio player
audioSourceBlindSGMSE.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-blind/sgmse/" + selectedAudioBlind + ".wav";
audioPlayerBlindSGMSE.load(); // Load the selected audio file using the audio player
// This will ensure that the audio players are updating their children source elements
// and reloading them
}
}
</script>
<br>
<h3>
Evaluation on real-world data
</h3>
<p>
This demo showcases the denoising capabilities of SGMSE+ <span class="reference" data-ref="sgmse"></span> trained using the EARS-WHAM dataset.
The red frame represents the noisy input audio, while the green frame indicates the enhanced, noise-reduced output.
</p>
<div class="video-container">
<iframe src="https://www.youtube.com/embed/H5FiO0JxPK4" frameborder="0" allowfullscreen></iframe>
</div>
<br>
<h3>
Dereverberation (EARS-Reverb)
</h3>
<p>
For the task of dereverberation, we use real recorded room impulse responses (RIRs) from multiple public datasets
<span class="reference" data-ref="ace air arni brudex dechorate detmoldsrir palimpsest"></span>. We generate
reverberant speech by convolving the clean speech with the RIR. More details can be found in the
<a href="https://arxiv.org/abs/2406.06185" target="_blank">paper</a>.
</p>
<h4>
Results
</h4>
<div class="table-container">
<table>
<caption style="caption-side: bottom; text-align: left; padding: 8px; font-style: italic;">
<strong>Table 4: Results on EARS-Reverb.</strong> Values indicate the mean of the metrics over the test set. The
best results are highlighted in bold.
</caption>
<thead>
<tr style="border-top: 2px solid black; border-bottom: 2px solid black;">
<th></th>
<th style="text-align: right;">POLQA</th>
<th style="text-align: right;">SI-SDR</th>
<th style="text-align: right;">PESQ</th>
<th style="text-align: right;">ESTOI</th>
<th style="text-align: right;">MOS Reverb</th>
</tr>
</thead>
<tbody>
<tr>
<td>Reverberant</td>
<td style="text-align: right;">1.75</td>
<td style="text-align: right;">-16.17</td>
<td style="text-align: right;">1.48</td>
<td style="text-align: right;">0.52</td>
<td style="text-align: right;">2.99</td>
</tr>
<tr style="border-bottom: 2px solid black;">
<td>SGMSE+ <span class="reference" data-ref="sgmse"></span></td>
<td style="text-align: right;"><strong>3.61</strong></td>
<td style="text-align: right;"><strong>5.79</strong></td>
<td style="text-align: right;"><strong>3.03</strong></td>
<td style="text-align: right;"><strong>0.85</strong></td>
<td style="text-align: right;"><strong>4.73</strong></td>
</tr>
</tbody>
</table>
</div>
<h4>
Audio Examples
</h4>
<p>
Here we present audio examples for the dereverberation task. Below we show the reverberant input, processed files
for SGMSE+ <span class="reference" data-ref="sgmse"></span>, and the clean ground truth.
</p>
<p>
Select an audio file:
<select id="audioSelectReverb" onchange="playAudioReverb()">
<option value="p102/00328_0.64">p102/00328_0.64</option>
<option value="p103/00296_0.61">p103/00296_0.61</option>
<option value="p104/00226_0.76">p104/00226_0.76</option>
<option value="p105/00067_0.43">p105/00067_0.43</option>
<option value="p106/00374_0.22">p106/00374_0.22</option>
<option value="p107/00111_0.30">p107/00111_0.30</option>
<option value="p102/00086_0.25">p102/00086_0.25</option>
<option value="p103/00259_0.22">p103/00259_0.22</option>
<option value="p104/00423_0.90">p104/00423_0.90</option>
<option value="p105/00574_0.64">p105/00574_0.64</option>
<option value="p106/00253_0.63">p106/00253_0.63</option>
<option value="p107/00097_0.26">p107/00097_0.26</option>
<option value="p102/00007_1.79">p102/00007_1.79</option>
<option value="p103/00667_0.92">p103/00667_0.92</option>
<option value="p104/00509_0.60">p104/00509_0.60</option>
<option value="p105/00336_0.51">p105/00336_0.51</option>
<option value="p106/00251_0.30">p106/00251_0.30</option>
<option value="p107/00140_0.41">p107/00140_0.41</option>
<option value="p102/00447_0.20">p102/00447_0.20</option>
<option value="p103/00516_0.70">p103/00516_0.70</option>
<option value="p104/00529_0.23">p104/00529_0.23</option>
<option value="p105/00368_0.82">p105/00368_0.82</option>
<option value="p106/00352_1.04">p106/00352_1.04</option>
<option value="p107/00682_0.56">p107/00682_0.56</option>
<option value="p102/00537_0.30">p102/00537_0.30</option>
<option value="p103/00479_0.68">p103/00479_0.68</option>
<option value="p104/00292_1.66">p104/00292_1.66</option>
<option value="p105/00733_0.46">p105/00733_0.46</option>
<option value="p106/00028_0.30">p106/00028_0.30</option>
<option value="p107/00092_0.19">p107/00092_0.19</option>
</select>
</p>
<p>
Reverberant: <br>
<audio id="audioPlayerReverb" controls>
<source id="audioSourceReverb" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-reverb/noisy/p102/00328_0.64.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
SGMSE+ <span class="reference" data-ref="sgmse"></span>:<br>
<audio id="audioPlayerReverbSGMSE" controls>
<source id="audioSourceReverbSGMSE" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-reverb/sgmse/p102/00328_0.64.wav" type="audio/wav">
Your browser does not support the audio element.
</audio><br>
Clean: <br>
<audio id="audioPlayerReverbClean" controls>
<source id="audioSourceReverbClean" src="https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-reverb/clean/p102/00328_0.64.wav" type="audio/wav">
Your browser does not support the audio element.
</audio>
</p>
<script>
function playAudioReverb() {
var audioPlayerReverb = document.getElementById('audioPlayerReverb');
var audioPlayerReverbSGMSE = document.getElementById('audioPlayerReverbSGMSE');
var audioPlayerReverbClean = document.getElementById('audioPlayerReverbClean');
var audioSourceReverb = document.getElementById('audioSourceReverb');
var audioSourceReverbSGMSE = document.getElementById('audioSourceReverbSGMSE');
var audioSourceReverbClean = document.getElementById('audioSourceReverbClean');
var selectedAudioReverb = document.getElementById('audioSelectReverb').value;
if (selectedAudioReverb) {
// Set the source of the source elements, not the audio players
audioSourceReverb.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-reverb/noisy/" + selectedAudioReverb + ".wav";
audioPlayerReverb.load(); // Load the selected audio file using the audio player
audioSourceReverbSGMSE.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-reverb/sgmse/" + selectedAudioReverb + ".wav";
audioPlayerReverbSGMSE.load(); // Load the selected audio file using the audio player
audioSourceReverbClean.src = "https://www2.informatik.uni-hamburg.de/sp/audio/publications/interspeech2024-ears/files/ears-reverb/clean/" + selectedAudioReverb + ".wav";
audioPlayerReverbClean.load(); // Load the selected audio file using the audio player
// This will ensure that the audio players are updating their children source elements
// and reloading them
}
}
</script>
<br>
<h2 id="citation">
Citation
</h2>
<p>
If you use the dataset or any derivative of it, please cite our
<a href="https://arxiv.org/abs/2406.06185" target="_blank">paper</a>:
</p>
<div class="container">
<div class="inner-container" dir="auto" data-snippet-clipboard-copy-content="{% raw %}@inproceedings{richter2024ears,
title={{EARS}: An Anechoic Fullband Speech Dataset Benchmarked for Speech Enhancement and Dereverberation},
author={Julius Richter and Yi-Chiao Wu and Steven Krenn and Simon Welker and Bunlong Lay and Shinjii Watanabe and Alexander Richard and Timo Gerkmann},
booktitle={ISCA Interspeech},
pages={4873--4877},
year={2024}
}{% endraw %}">
<pre><code>{% raw %}@inproceedings{richter2024ears,
title={{EARS}: An Anechoic Fullband Speech Dataset Benchmarked for Speech Enhancement and Dereverberation},
author={Julius Richter and Yi-Chiao Wu and Steven Krenn and Simon Welker and Bunlong Lay and Shinjii Watanabe and Alexander Richard and Timo Gerkmann},
booktitle={ISCA Interspeech},
pages={4873--4877},
year={2024}
}{% endraw %}</code></pre>
</div>
<button id="copyButton" class="highlight">
<svg xmlns="http://www.w3.org/2000/svg" height="16" viewBox="0 0 16 16" width="16" class="octicon octicon-clippy">
<path d="M0 6.75C0 5.784.784 5 1.75 5h1.5a.75.75 0 0 1 0 1.5h-1.5a.25.25 0 0 0-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 0 0 .25-.25v-1.5a.75.75 0 0 1 1.5 0v1.5A1.75 1.75 0 0 1 9.25 16h-7.5A1.75 1.75 0 0 1 0 14.25Z"></path>
<path d="M5 1.75C5 .784 5.784 0 6.75 0h7.5C15.216 0 16 .784 16 1.75v7.5A1.75 1.75 0 0 1 14.25 11h-7.5A1.75 1.75 0 0 1 5 9.25Zm1.75-.25a.25.25 0 0 0-.25.25v7.5c0 .138.112.25.25.25h7.5a.25.25 0 0 0 .25-.25v-7.5a.25.25 0 0 0-.25-.25Z"></path>
</svg>
</button>
</div>
<script>
document.getElementById('copyButton').addEventListener('click', function() {
const codeContent = document.querySelector('div[data-snippet-clipboard-copy-content]').getAttribute('data-snippet-clipboard-copy-content');
navigator.clipboard.writeText(codeContent).then(function() {
}, function() {
alert('Failed to copy!');
});
});
</script>
<br>
<h2>
References
</h2>
<ol id="refList" style="list-style-type: none; padding: 0;">
<!-- JavaScript will populate this list -->
</ol>
<script>
document.addEventListener("DOMContentLoaded", function () {
const references = {
"librivox": "H. Dubey, A. Aazami, V. Gopal, B. Naderi, S. Braun, R. Cutler, H. Gamper, M. Golestaneh, and R. Aichner, “ICASSP 2023 deep noise suppression challenge,” in IEEE International Conference on Acoustics, Speech and Signal Processing, 2023.",
"librispeech": "V. Panayotov, G. Chen, D. Povey, and S. Khudanpur, “Librispeech: An ASR corpus based on public domain audio books,” in IEEE International Conference on Acoustics, Speech and Signal Processing, 2015, pp. 5206–5210.",
"ljspeech": "K. Ito and L. Johnson, “The LJ Speech Dataset,” 2017. [Online]. Available: https://keithito.com/LJ-Speech-Dataset/",
"timit": "J. S. Garofolo, “TIMIT acoustic phonetic continuous speech corpus,” Linguistic Data Consortium, 1993.",
"vctk": "J. Yamagishi, C. Veaux, and K. MacDonald, “CSTR VCTK corpus: English multi-speaker corpus for CSTR voice cloning toolkit (version 0.92),” 2019. [Online]. Available: https://datashare.ed.ac.uk/handle/10283/3443",
"wsj0": "J. S. Garofolo, D. Graff, D. Paul, and D. Pallett, “CSR-I (WSJ0) Complete - Linguistic Data Consortium,” 1993. [Online]. Available: https://catalog.ldc.upenn.edu/LDC93s6a",
"convtasnet": "Y. Luo and N. Mesgarani, “Conv-TasNet: Surpassing ideal time–frequency magnitude masking for speech separation,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 27, no. 8, pp. 1256–1266, 2019.",
"cdiffuse": "Y.-J. Lu, Z.-Q. Wang, S. Watanabe, A. Richard, C. Yu, and Y. Tsao, “Conditional diffusion probabilistic model for speech enhancement,” in IEEE International Conference on Acoustics, Speech and Signal Processing, 2022, pp. 7402–7406.",
"demucs": "S. Rouard, F. Massa, and A. Défossez, “Hybrid transformers for music source separation,” in IEEE International Conference on Acoustics, Speech and Signal Processing, 2023.",
"sgmse": "J. Richter, S. Welker, J.-M. Lemercier, B. Lay, and T. Gerkmann, “Speech enhancement and dereverberation with diffusion-based generative models,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 31, pp. 2351–2364, 2023.",
"wham": "G. Wichern, J. Antognini, M. Flynn, L. R. Zhu, E. McQuinn, D. Crow, E. Manilow, and J. L. Roux, “WHAM!: Extending speech separation to noisy environments,” in ISCA Interspeech, 2019, pp. 1368–1372.",
"ace": "J. Eaton, N. D. Gaubitch, A. H. Moore, and P. A. Naylor, “Estimation of room acoustic parameters: The ACE challenge,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 24, no. 10, pp. 1681–1693, 2016.",
"air": "M. Jeub, M. Schafer, and P. Vary, “A binaural room impulse response database for the evaluation of dereverberation algorithms,” in IEEE International Conference on Digital Signal Processing, 2009.",
"arni": "K. Prawda, S. J. Schlecht, and V. V ̈alim ̈aki, “Robust selection of clean swept-sine measurements in non-stationary noise,” The Journal of the Acoustical Society of America, vol. 151, no. 3, pp. 2117–2126, 2022.",
"brudex": "D. Fejgin, W. Middelberg, and S. Doclo, “BRUDEX database: Binaural room impulse responses with uniformly distributed external microphones,” in Proc. ITG Conference on Speech Communication, 2023, pp. 126–130.",
"dechorate": "D. D. Carlo, P. Tandeitnik, C. Foy, N. Bertin, A. Deleforge, and S. Gannot, “dEchorate: a calibrated room impulse response dataset for echo-aware signal processing,” EURASIP Journal on Audio, Speech, and Music Processing, 2021.",
"detmoldsrir": "S. V. Amengual Gari, B. Sahin, D. Eddy, and M. Kob, “Open database of spatial room impulse responses at Detmold university of music,” in Audio Engineering Society Convention 149, 2020.",
"palimpsest": "“A sonic Palimpsest: Revisiting Chatham historic dockyards.” [Online]. Available: https://research.kent.ac.uk/sonic-palimpsest/impulse-responses/"
};
// Collect keys in order of their appearance, without duplication
const referenceElements = document.querySelectorAll('.reference');
const orderedKeys = [];
referenceElements.forEach(element => {
// Split the keys and iterate over them
const keys = element.dataset.ref.split(' ');
keys.forEach(key => {
// Add key if it's not already in the list to maintain first appearance order
if (!orderedKeys.includes(key) && references[key]) {
orderedKeys.push(key);
}
});
});
let sortedReferences = {};
// Populate the sorted references object based on orderedKeys
orderedKeys.forEach(key => {
if (references[key]) {
sortedReferences[key] = references[key];
}
});
function changeId(element) {
// Change the id of the clicked element to the next one
key = element.id.split('-')[1];
elem = document.getElementById(`link-${key}`);
elem.href = `#${element.id}`;
}
const refList = document.getElementById('refList');
const refTotal = Object.keys(sortedReferences).length;
const maxWidth = `${Math.ceil(Math.log10(refTotal + 1)) * 10 + 5}px`;
Object.entries(sortedReferences).forEach(([key, ref], index) => {
const refNumber = index + 1;
const li = document.createElement('li');
li.id = `ref-${refNumber}`;
li.style.display = 'flex';
li.style.alignItems = 'baseline';
li.style.marginBottom = '5px';
const numberSpan = document.createElement('span');
numberSpan.style.fontWeight = 'regular';
numberSpan.style.minWidth = maxWidth;
numberSpan.style.textAlign = 'right';
numberSpan.style.marginRight = '10px';
numberSpan.style.whiteSpace = 'nowrap';