Hugging Face's logo

Datasets:
DimQ1
/
sortformer-diarization-test-set

Tasks:
Audio Classification
Modalities:
Audio
Formats:
soundfolder
Languages:
English
Size:
< 1K
Libraries:
Datasets
License:
Dataset card Data Studio Files
xet
 Community
1
DimQ1 commited on
Commit
92c3f79
·
verified ·
1 Parent(s): 7f2ba2f

Add 100 real speech samples from LibriSpeech test-clean for diarization testing

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +67 -0
  2. audio/ls_real_000.wav +3 -0
  3. audio/ls_real_001.wav +3 -0
  4. audio/ls_real_002.wav +3 -0
  5. audio/ls_real_003.wav +3 -0
  6. audio/ls_real_004.wav +3 -0
  7. audio/ls_real_005.wav +3 -0
  8. audio/ls_real_006.wav +3 -0
  9. audio/ls_real_007.wav +3 -0
  10. audio/ls_real_008.wav +3 -0
  11. audio/ls_real_009.wav +3 -0
  12. audio/ls_real_010.wav +3 -0
  13. audio/ls_real_011.wav +3 -0
  14. audio/ls_real_012.wav +3 -0
  15. audio/ls_real_013.wav +3 -0
  16. audio/ls_real_014.wav +3 -0
  17. audio/ls_real_015.wav +3 -0
  18. audio/ls_real_016.wav +3 -0
  19. audio/ls_real_017.wav +3 -0
  20. audio/ls_real_018.wav +3 -0
  21. audio/ls_real_019.wav +3 -0
  22. audio/ls_real_020.wav +3 -0
  23. audio/ls_real_021.wav +3 -0
  24. audio/ls_real_022.wav +3 -0
  25. audio/ls_real_023.wav +3 -0
  26. audio/ls_real_024.wav +3 -0
  27. audio/ls_real_025.wav +3 -0
  28. audio/ls_real_026.wav +3 -0
  29. audio/ls_real_027.wav +3 -0
  30. audio/ls_real_028.wav +3 -0
  31. audio/ls_real_029.wav +3 -0
  32. audio/ls_real_030.wav +3 -0
  33. audio/ls_real_031.wav +3 -0
  34. audio/ls_real_032.wav +3 -0
  35. audio/ls_real_033.wav +3 -0
  36. audio/ls_real_034.wav +3 -0
  37. audio/ls_real_035.wav +3 -0
  38. audio/ls_real_036.wav +3 -0
  39. audio/ls_real_037.wav +3 -0
  40. audio/ls_real_038.wav +3 -0
  41. audio/ls_real_039.wav +3 -0
  42. audio/ls_real_040.wav +3 -0
  43. audio/ls_real_041.wav +3 -0
  44. audio/ls_real_042.wav +3 -0
  45. audio/ls_real_043.wav +3 -0
  46. audio/ls_real_044.wav +3 -0
  47. audio/ls_real_045.wav +3 -0
  48. audio/ls_real_046.wav +3 -0
  49. audio/ls_real_047.wav +3 -0
  50. audio/ls_real_048.wav +3 -0
README.md ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ task_categories:
4
+ - audio-classification
5
+ - speaker-diarization
6
+ language:
7
+ - en
8
+ pretty_name: Sortformer Diarization Test Set
9
+ size_categories:
10
+ - <1K
11
+ ---
12
+
13
+ # Sortformer Diarization Test Set
14
+
15
+ 100 real speech samples extracted from LibriSpeech test-clean for speaker diarization testing and benchmarking with NVIDIA Sortformer 4spk-v2 ONNX models.
16
+
17
+ ## Dataset Description
18
+
19
+ - **Samples**: 100 audio files (WAV, 16 kHz mono)
20
+ - **Total size**: ~60 MB
21
+ - **Source**: [LibriSpeech ASR corpus](https://www.openslr.org/12) — test-clean subset
22
+ - **Speakers**: 40 unique speakers from LibriSpeech test-clean
23
+ - **Purpose**: Diarization model evaluation, regression testing, ONNX model validation
24
+
25
+ ## Usage with Sortformer ONNX
26
+
27
+ ```python
28
+ from huggingface_hub import snapshot_download
29
+ import soundfile as sf
30
+
31
+ # Download the test set
32
+ dataset_path = snapshot_download("DimQ1/sortformer-diarization-test-set")
33
+
34
+ # Load audio
35
+ audio, sr = sf.read(f"{dataset_path}/audio/ls_real_000.wav")
36
+ ```
37
+
38
+ ## Diarization Models
39
+
40
+ Compatible ONNX models available on HuggingFace:
41
+
42
+ | Model | Size | Speed | Repo |
43
+ |-------|------|-------|------|
44
+ | Sortformer FP32 | 470 MB | 16× real-time | `DimQ1/sortformer-4spk-v2-onnx-fp32-cpu` |
45
+ | Sortformer INT8 | 129 MB | 29× real-time | `DimQ1/sortformer-4spk-v2-onnx-int8-cpu` |
46
+ | Sortformer INT4 | 73 MB | 33× real-time | `DimQ1/sortformer-4spk-v2-onnx-int4-cpu` |
47
+
48
+ ## Ground Truth RTTM
49
+
50
+ Speaker diarization ground truth annotations are provided in `rttm/` directory (NIST RTTM format).
51
+
52
+ ## Similar Datasets
53
+
54
+ For larger-scale diarization training and evaluation:
55
+
56
+ | Dataset | Description | Source |
57
+ |---------|-------------|--------|
58
+ | **LibriSpeech** | 1000h English read speech | [openslr.org/12](https://www.openslr.org/12) |
59
+ | **VoxCeleb 1&2** | 7000+ celebrity speakers | [robots.ox.ac.uk/~vgg/data/voxceleb](https://www.robots.ox.ac.uk/~vgg/data/voxceleb/) |
60
+ | **AMI Corpus** | 100h meeting recordings | [groups.inf.ed.ac.uk/ami/corpus](https://groups.inf.ed.ac.uk/ami/corpus/) |
61
+ | **CALLHOME** | Multilingual telephone speech | [catalog.ldc.upenn.edu/LDC97S42](https://catalog.ldc.upenn.edu/LDC97S42) |
62
+ | **DIHARD III** | Challenging diarization benchmark | [dihardchallenge.github.io/dihard3](https://dihardchallenge.github.io/dihard3/) |
63
+ | **MUSAN** | Music/speech/noise for augmentation | [openslr.org/17](https://www.openslr.org/17/) |
64
+
65
+ ## License
66
+
67
+ Derived from LibriSpeech (CC BY 4.0). See [LibriSpeech license](https://www.openslr.org/12) for details.
audio/ls_real_000.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d866b557fb6d453df276ab37a0b6c128c7b0aa92863aa880953d4ad2947f5fc
3
+ size 916754
audio/ls_real_001.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc9ac3d9147689de0025f6f352b2284e27e612a57f0c6791a6dbd54e53833270
3
+ size 812784
audio/ls_real_002.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da96cc5675b73d779692dd5670271e339f61aba505f417ab25077bc6cc9f2204
3
+ size 353274
audio/ls_real_003.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:005699c3de6c0ed38e1fddf958443e6fed679153909d53be400c222377ec33f4
3
+ size 362700
audio/ls_real_004.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22e8fd79a2e72b6a0fa87cdab09d1ded673f6b248af4f19ad7bc95c43e6f2a14
3
+ size 471490
audio/ls_real_005.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12b055b395f94e56e355b77af7ef796576673bee8ea3368e74ccefcb03712358
3
+ size 336928
audio/ls_real_006.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a1268ba7d072a60ef194351790593bc13a90a871e9963f92902addd14c8429f
3
+ size 1051640
audio/ls_real_007.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0686098b2193c8a7c830e47428a89dd9dc66c971c2b98bd6a509c45d10902864
3
+ size 464878
audio/ls_real_008.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a7f81c6f85961fe0fb16fecb123a7ee16081f38f4e319bb9c73a37d16ffc834
3
+ size 318036
audio/ls_real_009.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d85db9315014a1ff1921fbdc87a010f0bd933bc1f0db17cda9ab0e9a2c88e06
3
+ size 464090
audio/ls_real_010.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dc9ee25d57c5bb03f0f30eaa323a04760f6c5244f89636c9361bc9f65ed2e66
3
+ size 675774
audio/ls_real_011.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c46b8af006722c86a82193749eab5d181ac4e38394055b46473df4fad4ce6c9
3
+ size 318350
audio/ls_real_012.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11dac988c93a33c4e0d67e3e30ebde889743f38bd7c73900869c5a11dadf7ea6
3
+ size 784282
audio/ls_real_013.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fc8dd8d6abdc426a38bb5fa1beddf01ea311f92059478fc28b3e93e7b015955
3
+ size 276824
audio/ls_real_014.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e6bb9a9742d5b6b4ceff153c20dd5a760e625360b5a43065de9c6cbc20d9732
3
+ size 407852
audio/ls_real_015.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42a2897cd70bbb31f63f9b92531eddefb608f4dca979e8ae3b550218f0bddbde
3
+ size 337546
audio/ls_real_016.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:696a3c6e5c3c9cd4b20bc872205f15bb3288c801aaed1267ba9287bc875bebc1
3
+ size 1149644
audio/ls_real_017.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f8258eced39f750b4c199b9ebbd8df93f83705d628dd64d530887350dedcce2
3
+ size 928060
audio/ls_real_018.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3fed7f5e17be6b50f3c53046a3a94082a09a85938e000fc25ecb97bed595f27
3
+ size 450764
audio/ls_real_019.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16fc510e2f898538c67b70dcbb8cd77d160f87fc1bf0a88db682bc8ce2fc4c8b
3
+ size 595156
audio/ls_real_020.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca1b769130015681133fe4045974cea2f39b839f7e1fea6e2590314a6afba549
3
+ size 506264
audio/ls_real_021.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39c47542e4ff9a83954c8fff120ca120411361720fb346f516c83644eb2cf158
3
+ size 694420
audio/ls_real_022.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:167534d3b593ecefdd4efd18f485d47008a93b1d7e22916b3a9e87aa913a5d64
3
+ size 735602
audio/ls_real_023.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a8c85c5276c0ac244a2b9cd0a7b38f4fcdd388d92884212a474a59d9c3d3b8f
3
+ size 442790
audio/ls_real_024.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f51e6ae62c88ce1237d084b5b0d21cb0fdafb18dfc645d2e1ecbb7adf66858cb
3
+ size 501410
audio/ls_real_025.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e28b09051752282bea9bc14e7ce69c84b5d9fb32f36cb86354fd57162587d9d
3
+ size 578398
audio/ls_real_026.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4784f5633547f190e0e20554adf06f45e29a4a1afb7873db39f90c50346b4a3e
3
+ size 826496
audio/ls_real_027.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:588bb07b4b0aa39317fd573c32c224ba6292fc5107d16697af8f5c870a48db16
3
+ size 477040
audio/ls_real_028.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c00d16d4f7d01ed6990f9bdf02f21f890228aa15f840877000409f5b0417e0c
3
+ size 859244
audio/ls_real_029.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:005b9e6bbffe3b18161fc4f472fbbda5a4c2e0f352fa13b91379cd5134dd0e2c
3
+ size 550360
audio/ls_real_030.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:869ff50daaf3466495c7959b8be66fec9b02a19176b648d944a019784b3c8f70
3
+ size 383496
audio/ls_real_031.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7c6d9dbe0401f5146543aa027a17a2886d84d9ff1dbcf87f7266ef2ac57383e
3
+ size 555554
audio/ls_real_032.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe9cdfe6c800d1d7d4b8651679cb86d12e14d94a31d586d909b3f61600306eba
3
+ size 826014
audio/ls_real_033.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aff7ed7ffaaa46e71bb9b147a10fad8e8a1f47e4e60ad976e5b78e98f96d4710
3
+ size 904784
audio/ls_real_034.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc1660fc232a2eba25ecdffd872ecdef2d531cff2b55b1fc7f317048ed8abfba
3
+ size 276190
audio/ls_real_035.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:073ac60d8b090c5e9b317c735a3862219d616a1b30cb46a2d0dc477fe7fe3f7d
3
+ size 325646
audio/ls_real_036.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aacf9a852f29395e617790e402a0f642daa404e6308d2b1202356ae766ce8299
3
+ size 740500
audio/ls_real_037.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1f13dda4a80e9f7848ad71c3b72f90f8223537d96e514327e0b28781aa7623e
3
+ size 996430
audio/ls_real_038.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2dcd982a2b4002d565fff3bfe287baa74325d2a1457e8e86e7933398bebbb38
3
+ size 670644
audio/ls_real_039.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9fa4f1e05e04aa3d5f972aac5f29513866b0b33c49a2658a3588be5704f25bc
3
+ size 804918
audio/ls_real_040.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:458f5adafcacc4107ae962804fb3b5a866ad786e7388106f75a57a11882ee5b9
3
+ size 639464
audio/ls_real_041.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5577f10ff83c08dd1a1278761207c655abb716f79601327f31c6924baf6e5eb9
3
+ size 333030
audio/ls_real_042.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6dd1d675f31e2403bc3f4f9870b8e468369b02f3c31651a11440ae1cfef0dc9c
3
+ size 336312
audio/ls_real_043.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd6e22e0f9d863bead113953ec34ca805c45ce897db05b0ea3ed0e1bdcbe1700
3
+ size 770886
audio/ls_real_044.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a635ef33ead309db14f761482af2473095654e8a31a8b46349c3ada062347885
3
+ size 537320
audio/ls_real_045.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2b1a21495ca17246c0b47c58f3e731a4e54b8d4d5db4247417db81e6a289f69
3
+ size 887656
audio/ls_real_046.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3c591e86d4e4132576e834ec037b436beace17328c2e99a7757c327aa4748fe
3
+ size 340062
audio/ls_real_047.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45e52622b27dfe1ebd51131414857d7b38ef577db565da5f0a72130f5dcff08d
3
+ size 1215474
audio/ls_real_048.wav ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a12bb517c97385e8aa7eb1faa136eb49b76ea7603ab29c5d8c6031c8d2e3cae9
3
+ size 1170980