system HF staff commited on
Commit
cf54b2f
1 Parent(s): cb0e6fc

Update files from the datasets library (from 1.13.0)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.13.0

Files changed (3) hide show
  1. README.md +76 -17
  2. dataset_infos.json +1 -1
  3. superb.py +207 -7
README.md CHANGED
@@ -87,6 +87,37 @@ Automatic Speech Recognition (ASR) transcribes utterances into words. While PR a
87
 
88
  Keyword Spotting (KS) detects preregistered keywords by classifying utterances into a predefined set of words. The task is usually performed on-device for the fast response time. Thus, accuracy, model size, and inference time are all crucial. SUPERB uses the widely used [Speech Commands dataset v1.0](https://www.tensorflow.org/datasets/catalog/speech_commands) for the task. The dataset consists of ten classes of keywords, a class for silence, and an unknown class to include the false positive. The evaluation metric is accuracy (ACC)
89
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
  #### qbe
91
 
92
  Query by Example Spoken Term Detection (QbE) detects a spoken term (query) in an audio database (documents) by binary discriminating a given pair of query and document into a match or not. The English subset in [QUESST 2014 challenge](https://github.com/s3prl/s3prl/tree/master/downstream#qbe-query-by-example-spoken-term-detection) is adopted since we focus on investigating English as the first step. The evaluation metric is maximum term weighted value (MTWV) which balances misses and false alarms.
@@ -189,7 +220,7 @@ An example from each split looks like:
189
  ```python
190
  {
191
  'file': '/path/yes/af7a8296_nohash_1.wav',
192
- 'label': 'yes'
193
  }
194
  ```
195
 
@@ -200,8 +231,16 @@ An example from each split looks like:
200
 
201
  #### ic
202
 
203
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
204
-
 
 
 
 
 
 
 
 
205
 
206
  #### sf
207
 
@@ -210,8 +249,12 @@ An example from each split looks like:
210
 
211
  #### si
212
 
213
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
214
-
 
 
 
 
215
 
216
  #### asv
217
 
@@ -260,17 +303,24 @@ An example from each split looks like:
260
  #### ks
261
 
262
  - `file` (`string`): Path to the WAV audio file.
263
- - `label` (`string`): Label of the spoken command.
 
264
 
265
  #### qbe
266
 
267
  [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
268
 
269
-
270
  #### ic
271
 
272
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
273
-
 
 
 
 
 
 
 
274
 
275
  #### sf
276
 
@@ -279,8 +329,9 @@ An example from each split looks like:
279
 
280
  #### si
281
 
282
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
283
-
 
284
 
285
  #### asv
286
 
@@ -301,7 +352,9 @@ The data fields in all splits are:
301
 
302
  #### er
303
 
304
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
 
 
305
 
306
  ### Data Splits
307
 
@@ -329,8 +382,9 @@ The data fields in all splits are:
329
 
330
  #### ic
331
 
332
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
333
-
 
334
 
335
  #### sf
336
 
@@ -339,8 +393,9 @@ The data fields in all splits are:
339
 
340
  #### si
341
 
342
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
343
-
 
344
 
345
  #### asv
346
 
@@ -357,7 +412,11 @@ The data is split into "train", "dev" and "test" sets, each containing the follo
357
 
358
  #### er
359
 
360
- [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
 
 
 
 
361
 
362
  ## Dataset Creation
363
 
 
87
 
88
  Keyword Spotting (KS) detects preregistered keywords by classifying utterances into a predefined set of words. The task is usually performed on-device for the fast response time. Thus, accuracy, model size, and inference time are all crucial. SUPERB uses the widely used [Speech Commands dataset v1.0](https://www.tensorflow.org/datasets/catalog/speech_commands) for the task. The dataset consists of ten classes of keywords, a class for silence, and an unknown class to include the false positive. The evaluation metric is accuracy (ACC)
89
 
90
+ ##### Example of usage:
91
+
92
+ Use these auxillary functions to:
93
+ - load the audio file into an audio data array
94
+ - sample from long `_silence_` audio clips
95
+
96
+ For other examples of handling long `_silence_` clips see the [S3PRL](https://github.com/s3prl/s3prl/blob/099ce807a6ffa6bf2482ceecfcaf83dea23da355/s3prl/downstream/speech_commands/dataset.py#L80)
97
+ or [TFDS](https://github.com/tensorflow/datasets/blob/6b8cfdb7c3c0a04e731caaa8660ce948d0a67b1e/tensorflow_datasets/audio/speech_commands.py#L143) implementations.
98
+
99
+ ```python
100
+ def map_to_array(example):
101
+ import soundfile as sf
102
+
103
+ speech_array, sample_rate = sf.read(example["file"])
104
+ example["speech"] = speech_array
105
+ example["sample_rate"] = sample_rate
106
+ return example
107
+
108
+
109
+ def sample_noise(example):
110
+ # Use this function to extract random 1 sec slices of each _silence_ utterance,
111
+ # e.g. inside `torch.utils.data.Dataset.__getitem__()`
112
+ from random import randint
113
+
114
+ if example["label"] == "_silence_":
115
+ random_offset = randint(0, len(example["speech"]) - example["sample_rate"] - 1)
116
+ example["speech"] = example["speech"][random_offset : random_offset + example["sample_rate"]]
117
+
118
+ return example
119
+ ```
120
+
121
  #### qbe
122
 
123
  Query by Example Spoken Term Detection (QbE) detects a spoken term (query) in an audio database (documents) by binary discriminating a given pair of query and document into a match or not. The English subset in [QUESST 2014 challenge](https://github.com/s3prl/s3prl/tree/master/downstream#qbe-query-by-example-spoken-term-detection) is adopted since we focus on investigating English as the first step. The evaluation metric is maximum term weighted value (MTWV) which balances misses and false alarms.
 
220
  ```python
221
  {
222
  'file': '/path/yes/af7a8296_nohash_1.wav',
223
+ 'label': 0 # 'yes'
224
  }
225
  ```
226
 
 
231
 
232
  #### ic
233
 
234
+ ```python
235
+ {
236
+ 'file': "/path/wavs/speakers/2BqVo8kVB2Skwgyb/063aa8f0-4479-11e9-a9a5-5dbec3b8816a.wav",
237
+ 'speaker_id': '2BqVo8kVB2Skwgyb',
238
+ 'text': 'Turn the bedroom lights off',
239
+ 'action': 3, # 'deactivate'
240
+ 'object': 7, # 'lights'
241
+ 'location': 0 # 'bedroom'
242
+ }
243
+ ```
244
 
245
  #### sf
246
 
 
249
 
250
  #### si
251
 
252
+ ```python
253
+ {
254
+ 'file': '/path/wav/id10003/na8-QEFmj44/00003.wav',
255
+ 'label': 2 # 'id10003'
256
+ }
257
+ ```
258
 
259
  #### asv
260
 
 
303
  #### ks
304
 
305
  - `file` (`string`): Path to the WAV audio file.
306
+ - `label` (`ClassLabel`): Label of the spoken command. Possible values:
307
+ - `0: "yes", 1: "no", 2: "up", 3: "down", 4: "left", 5: "right", 6: "on", 7: "off", 8: "stop", 9: "go", 10: "_silence_", 11: "_unknown_"`
308
 
309
  #### qbe
310
 
311
  [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
312
 
 
313
  #### ic
314
 
315
+ - `file` (`string`): Path to the WAV audio file.
316
+ - `speaker_id` (`string`): ID of the speaker.
317
+ - `text` (`string`): Transcription of the spoken command.
318
+ - `action` (`ClassLabel`): Label of the command's action. Possible values:
319
+ - `0: "activate", 1: "bring", 2: "change language", 3: "deactivate", 4: "decrease", 5: "increase"`
320
+ - `object` (`ClassLabel`): Label of the command's object. Possible values:
321
+ - `0: "Chinese", 1: "English", 2: "German", 3: "Korean", 4: "heat", 5: "juice", 6: "lamp", 7: "lights", 8: "music", 9: "newspaper", 10: "none", 11: "shoes", 12: "socks", 13: "volume"`
322
+ - `location` (`ClassLabel`): Label of the command's location. Possible values:
323
+ - `0: "bedroom", 1: "kitchen", 2: "none", 3: "washroom"`
324
 
325
  #### sf
326
 
 
329
 
330
  #### si
331
 
332
+ - `file` (`string`): Path to the WAV audio file.
333
+ - `label` (`ClassLabel`): Label (ID) of the speaker. Possible values:
334
+ - `0: "id10001", 1: "id10002", 2: "id10003", ..., 1250: "id11251"`
335
 
336
  #### asv
337
 
 
352
 
353
  #### er
354
 
355
+ - `file` (`string`): Path to the WAV audio file.
356
+ - `label` (`ClassLabel`): Label of the speech emotion. Possible values:
357
+ - `0: "neu", 1: "hap", 2: "ang", 3: "sad"`
358
 
359
  ### Data Splits
360
 
 
382
 
383
  #### ic
384
 
385
+ | | train | validation | test |
386
+ |----|------:|-----------:|-----:|
387
+ | ic | 23132 | 3118 | 3793 |
388
 
389
  #### sf
390
 
 
393
 
394
  #### si
395
 
396
+ | | train | validation | test |
397
+ |----|-------:|-----------:|-----:|
398
+ | si | 138361 | 6904 | 8251 |
399
 
400
  #### asv
401
 
 
412
 
413
  #### er
414
 
415
+ The data is split into 5 sets intended for 5-fold cross-validation:
416
+
417
+ | | session1 | session2 | session3 | session4 | session5 |
418
+ |----|---------:|---------:|---------:|---------:|---------:|
419
+ | er | 1085 | 1023 | 1151 | 1031 | 1241 |
420
 
421
  ## Dataset Creation
422
 
dataset_infos.json CHANGED
@@ -1 +1 @@
1
- {"asr": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .flac format and is not converted to a float32 array. To\nconvert, the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "http://www.openslr.org/12", "license": "", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}, "speaker_id": {"dtype": "int64", "id": null, "_type": "Value"}, "chapter_id": {"dtype": "int64", "id": null, "_type": "Value"}, "id": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "text"}, "task_templates": [{"task": "automatic-speech-recognition", "audio_file_path_column": "file", "transcription_column": "text"}], "builder_name": "superb", "config_name": "asr", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 11852430, "num_examples": 28539, "dataset_name": "superb"}, "validation": {"name": "validation", "num_bytes": 897213, "num_examples": 2703, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 871234, "num_examples": 2620, "dataset_name": "superb"}}, "download_checksums": {"http://www.openslr.org/resources/12/dev-clean.tar.gz": {"num_bytes": 337926286, "checksum": "76f87d090650617fca0cac8f88b9416e0ebf80350acb97b343a85fa903728ab3"}, "http://www.openslr.org/resources/12/test-clean.tar.gz": {"num_bytes": 346663984, "checksum": "39fde525e59672dc6d1551919b1478f724438a95aa55f874b576be21967e6c23"}, "http://www.openslr.org/resources/12/train-clean-100.tar.gz": {"num_bytes": 6387309499, "checksum": "d4ddd1d5a6ab303066f14971d768ee43278a5f2a0aa43dc716b0e64ecbbbf6e2"}}, "download_size": 7071899769, "post_processing_size": null, "dataset_size": 13620877, "size_in_bytes": 7085520646}, "sd": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .flac format and is not converted to a float32 array. To\nconvert, the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "https://github.com/ftshijt/LibriMix", "license": "", "features": {"record_id": {"dtype": "string", "id": null, "_type": "Value"}, "file": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int64", "id": null, "_type": "Value"}, "end": {"dtype": "int64", "id": null, "_type": "Value"}, "speakers": [{"speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int64", "id": null, "_type": "Value"}, "end": {"dtype": "int64", "id": null, "_type": "Value"}}]}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "superb", "config_name": "sd", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 4622013, "num_examples": 13901, "dataset_name": "superb"}, "dev": {"name": "dev", "num_bytes": 860472, "num_examples": 3014, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 847803, "num_examples": 3002, "dataset_name": "superb"}}, "download_checksums": {"https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/reco2dur": {"num_bytes": 540906, "checksum": "879dca4b1108c93bd86df879463fca15a4de42a0f95a7e6987138dc6029b5554"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/segments": {"num_bytes": 5723993, "checksum": "f19cb0ecc342f8d2cd855118879a111822d7cf55fcd078ef156f5147233a8e11"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/utt2spk": {"num_bytes": 3165995, "checksum": "a4295726caf05d72f5ad24706180b9dbccffe6c0c2fc0128ca4b02b7b828a28a"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/wav.zip": {"num_bytes": 5706733518, "checksum": "4231070427ffbc9b3bddae874dba32f3985a0db0b0feb4dfa29ed4d1d11bf41b"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/reco2dur": {"num_bytes": 115918, "checksum": "a30fd59ad01db0315a82cad7a64baea009e6c2bcdfb6b2501bc8873ede72de06"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/segments": {"num_bytes": 673006, "checksum": "2b977917e7ab9feec03afb4fd6a4662df90e48dbcc42977a4b9c89c8d40432ee"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/utt2spk": {"num_bytes": 374794, "checksum": "9f47a7bed76e7a03e57d66ba9cc5f57d85d91f748d0b1eb20301d09e6c24cd20"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/wav.zip": {"num_bytes": 765594100, "checksum": "e28b3422ce59e2a5273be924e6ed6b8f115c0983db1997e56441973c27ee1cd8"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/reco2dur": {"num_bytes": 113357, "checksum": "6e013d917015031e2f1383871b52dfc1122e7b16cdee53bd8e5e0a7fbc57e406"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/segments": {"num_bytes": 650742, "checksum": "92f8de0f56c55a34e9111542c24ea13f2d2efaf9ebe64af31250cadab020f987"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/utt2spk": {"num_bytes": 361548, "checksum": "19dcb558aa886f0d553d8d9b8735ea1998b83e96d5245e5511cb732c84625ffd"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/wav.zip": {"num_bytes": 706322334, "checksum": "9c8ee97d3068759c0101bf88684abab77183374dbb3bb40f7c0b25d385992ea6"}}, "download_size": 7190370211, "post_processing_size": null, "dataset_size": 6330288, "size_in_bytes": 7196700499}, "ks": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .flac format and is not converted to a float32 array. To\nconvert, the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "https://www.tensorflow.org/datasets/catalog/speech_commands", "license": "", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 12, "names": ["yes", "no", "up", "down", "left", "right", "on", "off", "stop", "go", "_silence_", "_unknown_"], "names_file": null, "id": null, "_type": "ClassLabel"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "label"}, "task_templates": null, "builder_name": "superb", "config_name": "ks", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 8467781, "num_examples": 51094, "dataset_name": "superb"}, "validation": {"name": "validation", "num_bytes": 1126476, "num_examples": 6798, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 510619, "num_examples": 3081, "dataset_name": "superb"}}, "download_checksums": {"http://download.tensorflow.org/data/speech_commands_v0.01.tar.gz": {"num_bytes": 1489096277, "checksum": "743935421bb51cccdb6bdd152e04c5c70274e935c82119ad7faeec31780d811d"}, "http://download.tensorflow.org/data/speech_commands_test_set_v0.01.tar.gz": {"num_bytes": 71271436, "checksum": "baa084f6b62c91de660ff0588ae4dfc4e4d534aa99ac0e5f406cba75836cbd00"}}, "download_size": 1560367713, "post_processing_size": null, "dataset_size": 10104876, "size_in_bytes": 1570472589}}
 
1
+ {"asr": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .wav format and is not converted to a float32 array. To\nconvert the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "http://www.openslr.org/12", "license": "", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}, "speaker_id": {"dtype": "int64", "id": null, "_type": "Value"}, "chapter_id": {"dtype": "int64", "id": null, "_type": "Value"}, "id": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "text"}, "task_templates": [{"task": "automatic-speech-recognition", "audio_file_path_column": "file", "transcription_column": "text"}], "builder_name": "superb", "config_name": "asr", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 11852430, "num_examples": 28539, "dataset_name": "superb"}, "validation": {"name": "validation", "num_bytes": 897213, "num_examples": 2703, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 871234, "num_examples": 2620, "dataset_name": "superb"}}, "download_checksums": {"http://www.openslr.org/resources/12/dev-clean.tar.gz": {"num_bytes": 337926286, "checksum": "76f87d090650617fca0cac8f88b9416e0ebf80350acb97b343a85fa903728ab3"}, "http://www.openslr.org/resources/12/test-clean.tar.gz": {"num_bytes": 346663984, "checksum": "39fde525e59672dc6d1551919b1478f724438a95aa55f874b576be21967e6c23"}, "http://www.openslr.org/resources/12/train-clean-100.tar.gz": {"num_bytes": 6387309499, "checksum": "d4ddd1d5a6ab303066f14971d768ee43278a5f2a0aa43dc716b0e64ecbbbf6e2"}}, "download_size": 7071899769, "post_processing_size": null, "dataset_size": 13620877, "size_in_bytes": 7085520646}, "sd": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .flac format and is not converted to a float32 array. To\nconvert, the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "https://github.com/ftshijt/LibriMix", "license": "", "features": {"record_id": {"dtype": "string", "id": null, "_type": "Value"}, "file": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int64", "id": null, "_type": "Value"}, "end": {"dtype": "int64", "id": null, "_type": "Value"}, "speakers": [{"speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int64", "id": null, "_type": "Value"}, "end": {"dtype": "int64", "id": null, "_type": "Value"}}]}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "superb", "config_name": "sd", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 4622013, "num_examples": 13901, "dataset_name": "superb"}, "dev": {"name": "dev", "num_bytes": 860472, "num_examples": 3014, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 847803, "num_examples": 3002, "dataset_name": "superb"}}, "download_checksums": {"https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/reco2dur": {"num_bytes": 540906, "checksum": "879dca4b1108c93bd86df879463fca15a4de42a0f95a7e6987138dc6029b5554"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/segments": {"num_bytes": 5723993, "checksum": "f19cb0ecc342f8d2cd855118879a111822d7cf55fcd078ef156f5147233a8e11"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/utt2spk": {"num_bytes": 3165995, "checksum": "a4295726caf05d72f5ad24706180b9dbccffe6c0c2fc0128ca4b02b7b828a28a"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/train/wav.zip": {"num_bytes": 5706733518, "checksum": "4231070427ffbc9b3bddae874dba32f3985a0db0b0feb4dfa29ed4d1d11bf41b"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/reco2dur": {"num_bytes": 115918, "checksum": "a30fd59ad01db0315a82cad7a64baea009e6c2bcdfb6b2501bc8873ede72de06"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/segments": {"num_bytes": 673006, "checksum": "2b977917e7ab9feec03afb4fd6a4662df90e48dbcc42977a4b9c89c8d40432ee"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/utt2spk": {"num_bytes": 374794, "checksum": "9f47a7bed76e7a03e57d66ba9cc5f57d85d91f748d0b1eb20301d09e6c24cd20"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/dev/wav.zip": {"num_bytes": 765594100, "checksum": "e28b3422ce59e2a5273be924e6ed6b8f115c0983db1997e56441973c27ee1cd8"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/reco2dur": {"num_bytes": 113357, "checksum": "6e013d917015031e2f1383871b52dfc1122e7b16cdee53bd8e5e0a7fbc57e406"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/segments": {"num_bytes": 650742, "checksum": "92f8de0f56c55a34e9111542c24ea13f2d2efaf9ebe64af31250cadab020f987"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/utt2spk": {"num_bytes": 361548, "checksum": "19dcb558aa886f0d553d8d9b8735ea1998b83e96d5245e5511cb732c84625ffd"}, "https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/test/wav.zip": {"num_bytes": 706322334, "checksum": "9c8ee97d3068759c0101bf88684abab77183374dbb3bb40f7c0b25d385992ea6"}}, "download_size": 7190370211, "post_processing_size": null, "dataset_size": 6330288, "size_in_bytes": 7196700499}, "ks": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .wav format and is not converted to a float32 array. To\nconvert the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "https://www.tensorflow.org/datasets/catalog/speech_commands", "license": "", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 12, "names": ["yes", "no", "up", "down", "left", "right", "on", "off", "stop", "go", "_silence_", "_unknown_"], "names_file": null, "id": null, "_type": "ClassLabel"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "label"}, "task_templates": null, "builder_name": "superb", "config_name": "ks", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 8467781, "num_examples": 51094, "dataset_name": "superb"}, "validation": {"name": "validation", "num_bytes": 1126476, "num_examples": 6798, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 510619, "num_examples": 3081, "dataset_name": "superb"}}, "download_checksums": {"http://download.tensorflow.org/data/speech_commands_v0.01.tar.gz": {"num_bytes": 1489096277, "checksum": "743935421bb51cccdb6bdd152e04c5c70274e935c82119ad7faeec31780d811d"}, "http://download.tensorflow.org/data/speech_commands_test_set_v0.01.tar.gz": {"num_bytes": 71271436, "checksum": "baa084f6b62c91de660ff0588ae4dfc4e4d534aa99ac0e5f406cba75836cbd00"}}, "download_size": 1560367713, "post_processing_size": null, "dataset_size": 10104876, "size_in_bytes": 1570472589}, "ic": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .flac format and is not converted to a float32 array. To\nconvert, the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "https://fluent.ai/fluent-speech-commands-a-dataset-for-spoken-language-understanding-research/", "license": "", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}, "action": {"num_classes": 6, "names": ["activate", "bring", "change language", "deactivate", "decrease", "increase"], "names_file": null, "id": null, "_type": "ClassLabel"}, "object": {"num_classes": 14, "names": ["Chinese", "English", "German", "Korean", "heat", "juice", "lamp", "lights", "music", "newspaper", "none", "shoes", "socks", "volume"], "names_file": null, "id": null, "_type": "ClassLabel"}, "location": {"num_classes": 4, "names": ["bedroom", "kitchen", "none", "washroom"], "names_file": null, "id": null, "_type": "ClassLabel"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "superb", "config_name": "ic", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 7071466, "num_examples": 23132, "dataset_name": "superb"}, "validation": {"name": "validation", "num_bytes": 953622, "num_examples": 3118, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 1158347, "num_examples": 3793, "dataset_name": "superb"}}, "download_checksums": {"http://fluent.ai:2052/jf8398hf30f0381738rucj3828chfdnchs.tar.gz": {"num_bytes": 1544093324, "checksum": "4376699f7daf134a9fa57a1d880ffcaaf94a3e2551ba0b40ad894d7abb71aacb"}}, "download_size": 1544093324, "post_processing_size": null, "dataset_size": 9183435, "size_in_bytes": 1553276759}, "si": {"description": "Self-supervised learning (SSL) has proven vital for advancing research in\nnatural language processing (NLP) and computer vision (CV). The paradigm\npretrains a shared model on large volumes of unlabeled data and achieves\nstate-of-the-art (SOTA) for various tasks with minimal adaptation. However, the\nspeech processing community lacks a similar setup to systematically explore the\nparadigm. To bridge this gap, we introduce Speech processing Universal\nPERformance Benchmark (SUPERB). SUPERB is a leaderboard to benchmark the\nperformance of a shared model across a wide range of speech processing tasks\nwith minimal architecture changes and labeled data. Among multiple usages of the\nshared model, we especially focus on extracting the representation learned from\nSSL due to its preferable re-usability. We present a simple framework to solve\nSUPERB tasks by learning task-specialized lightweight prediction heads on top of\nthe frozen shared model. Our results demonstrate that the framework is promising\nas SSL representations show competitive generalizability and accessibility\nacross SUPERB tasks. We release SUPERB as a challenge with a leaderboard and a\nbenchmark toolkit to fuel the research in representation learning and general\nspeech processing.\n\nNote that in order to limit the required storage for preparing this dataset, the\naudio is stored in the .flac format and is not converted to a float32 array. To\nconvert, the audio file to a float32 array, please make use of the `.map()`\nfunction as follows:\n\n\n```python\nimport soundfile as sf\n\ndef map_to_array(batch):\n speech_array, _ = sf.read(batch[\"file\"])\n batch[\"speech\"] = speech_array\n return batch\n\ndataset = dataset.map(map_to_array, remove_columns=[\"file\"])\n```\n", "citation": "@article{DBLP:journals/corr/abs-2105-01051,\n author = {Shu{-}Wen Yang and\n Po{-}Han Chi and\n Yung{-}Sung Chuang and\n Cheng{-}I Jeff Lai and\n Kushal Lakhotia and\n Yist Y. Lin and\n Andy T. Liu and\n Jiatong Shi and\n Xuankai Chang and\n Guan{-}Ting Lin and\n Tzu{-}Hsien Huang and\n Wei{-}Cheng Tseng and\n Ko{-}tik Lee and\n Da{-}Rong Liu and\n Zili Huang and\n Shuyan Dong and\n Shang{-}Wen Li and\n Shinji Watanabe and\n Abdelrahman Mohamed and\n Hung{-}yi Lee},\n title = {{SUPERB:} Speech processing Universal PERformance Benchmark},\n journal = {CoRR},\n volume = {abs/2105.01051},\n year = {2021},\n url = {https://arxiv.org/abs/2105.01051},\n archivePrefix = {arXiv},\n eprint = {2105.01051},\n timestamp = {Thu, 01 Jul 2021 13:30:22 +0200},\n biburl = {https://dblp.org/rec/journals/corr/abs-2105-01051.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n", "homepage": "https://www.robots.ox.ac.uk/~vgg/data/voxceleb/vox1.html", "license": "", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 1251, "names": ["id10001", "id10002", "id10003", "id10004", "id10005", "id10006", "id10007", "id10008", "id10009", "id10010", "id10011", "id10012", "id10013", "id10014", "id10015", "id10016", "id10017", "id10018", "id10019", "id10020", "id10021", "id10022", "id10023", "id10024", "id10025", "id10026", "id10027", "id10028", "id10029", "id10030", "id10031", "id10032", "id10033", "id10034", "id10035", "id10036", "id10037", "id10038", "id10039", "id10040", "id10041", "id10042", "id10043", "id10044", "id10045", "id10046", "id10047", "id10048", "id10049", "id10050", "id10051", "id10052", "id10053", "id10054", "id10055", "id10056", "id10057", "id10058", "id10059", "id10060", "id10061", "id10062", "id10063", "id10064", "id10065", "id10066", "id10067", "id10068", "id10069", "id10070", "id10071", "id10072", "id10073", "id10074", "id10075", "id10076", "id10077", "id10078", "id10079", "id10080", "id10081", "id10082", "id10083", "id10084", "id10085", "id10086", "id10087", "id10088", "id10089", "id10090", "id10091", "id10092", "id10093", "id10094", "id10095", "id10096", "id10097", "id10098", "id10099", "id10100", "id10101", "id10102", "id10103", "id10104", "id10105", "id10106", "id10107", "id10108", "id10109", "id10110", "id10111", "id10112", "id10113", "id10114", "id10115", "id10116", "id10117", "id10118", "id10119", "id10120", "id10121", "id10122", "id10123", "id10124", "id10125", "id10126", "id10127", "id10128", "id10129", "id10130", "id10131", "id10132", "id10133", "id10134", "id10135", "id10136", "id10137", "id10138", "id10139", "id10140", "id10141", "id10142", "id10143", "id10144", "id10145", "id10146", "id10147", "id10148", "id10149", "id10150", "id10151", "id10152", "id10153", "id10154", "id10155", "id10156", "id10157", "id10158", "id10159", "id10160", "id10161", "id10162", "id10163", "id10164", "id10165", "id10166", "id10167", "id10168", "id10169", "id10170", "id10171", "id10172", "id10173", "id10174", "id10175", "id10176", "id10177", "id10178", "id10179", "id10180", "id10181", "id10182", "id10183", "id10184", "id10185", "id10186", "id10187", "id10188", "id10189", "id10190", "id10191", "id10192", "id10193", "id10194", "id10195", "id10196", "id10197", "id10198", "id10199", "id10200", "id10201", "id10202", "id10203", "id10204", "id10205", "id10206", "id10207", "id10208", "id10209", "id10210", "id10211", "id10212", "id10213", "id10214", "id10215", "id10216", "id10217", "id10218", "id10219", "id10220", "id10221", "id10222", "id10223", "id10224", "id10225", "id10226", "id10227", "id10228", "id10229", "id10230", "id10231", "id10232", "id10233", "id10234", "id10235", "id10236", "id10237", "id10238", "id10239", "id10240", "id10241", "id10242", "id10243", "id10244", "id10245", "id10246", "id10247", "id10248", "id10249", "id10250", "id10251", "id10252", "id10253", "id10254", "id10255", "id10256", "id10257", "id10258", "id10259", "id10260", "id10261", "id10262", "id10263", "id10264", "id10265", "id10266", "id10267", "id10268", "id10269", "id10270", "id10271", "id10272", "id10273", "id10274", "id10275", "id10276", "id10277", "id10278", "id10279", "id10280", "id10281", "id10282", "id10283", "id10284", "id10285", "id10286", "id10287", "id10288", "id10289", "id10290", "id10291", "id10292", "id10293", "id10294", "id10295", "id10296", "id10297", "id10298", "id10299", "id10300", "id10301", "id10302", "id10303", "id10304", "id10305", "id10306", "id10307", "id10308", "id10309", "id10310", "id10311", "id10312", "id10313", "id10314", "id10315", "id10316", "id10317", "id10318", "id10319", "id10320", "id10321", "id10322", "id10323", "id10324", "id10325", "id10326", "id10327", "id10328", "id10329", "id10330", "id10331", "id10332", "id10333", "id10334", "id10335", "id10336", "id10337", "id10338", "id10339", "id10340", "id10341", "id10342", "id10343", "id10344", "id10345", "id10346", "id10347", "id10348", "id10349", "id10350", "id10351", "id10352", "id10353", "id10354", "id10355", "id10356", "id10357", "id10358", "id10359", "id10360", "id10361", "id10362", "id10363", "id10364", "id10365", "id10366", "id10367", "id10368", "id10369", "id10370", "id10371", "id10372", "id10373", "id10374", "id10375", "id10376", "id10377", "id10378", "id10379", "id10380", "id10381", "id10382", "id10383", "id10384", "id10385", "id10386", "id10387", "id10388", "id10389", "id10390", "id10391", "id10392", "id10393", "id10394", "id10395", "id10396", "id10397", "id10398", "id10399", "id10400", "id10401", "id10402", "id10403", "id10404", "id10405", "id10406", "id10407", "id10408", "id10409", "id10410", "id10411", "id10412", "id10413", "id10414", "id10415", "id10416", "id10417", "id10418", "id10419", "id10420", "id10421", "id10422", "id10423", "id10424", "id10425", "id10426", "id10427", "id10428", "id10429", "id10430", "id10431", "id10432", "id10433", "id10434", "id10435", "id10436", "id10437", "id10438", "id10439", "id10440", "id10441", "id10442", "id10443", "id10444", "id10445", "id10446", "id10447", "id10448", "id10449", "id10450", "id10451", "id10452", "id10453", "id10454", "id10455", "id10456", "id10457", "id10458", "id10459", "id10460", "id10461", "id10462", "id10463", "id10464", "id10465", "id10466", "id10467", "id10468", "id10469", "id10470", "id10471", "id10472", "id10473", "id10474", "id10475", "id10476", "id10477", "id10478", "id10479", "id10480", "id10481", "id10482", "id10483", "id10484", "id10485", "id10486", "id10487", "id10488", "id10489", "id10490", "id10491", "id10492", "id10493", "id10494", "id10495", "id10496", "id10497", "id10498", "id10499", "id10500", "id10501", "id10502", "id10503", "id10504", "id10505", "id10506", "id10507", "id10508", "id10509", "id10510", "id10511", "id10512", "id10513", "id10514", "id10515", "id10516", "id10517", "id10518", "id10519", "id10520", "id10521", "id10522", "id10523", "id10524", "id10525", "id10526", "id10527", "id10528", "id10529", "id10530", "id10531", "id10532", "id10533", "id10534", "id10535", "id10536", "id10537", "id10538", "id10539", "id10540", "id10541", "id10542", "id10543", "id10544", "id10545", "id10546", "id10547", "id10548", "id10549", "id10550", "id10551", "id10552", "id10553", "id10554", "id10555", "id10556", "id10557", "id10558", "id10559", "id10560", "id10561", "id10562", "id10563", "id10564", "id10565", "id10566", "id10567", "id10568", "id10569", "id10570", "id10571", "id10572", "id10573", "id10574", "id10575", "id10576", "id10577", "id10578", "id10579", "id10580", "id10581", "id10582", "id10583", "id10584", "id10585", "id10586", "id10587", "id10588", "id10589", "id10590", "id10591", "id10592", "id10593", "id10594", "id10595", "id10596", "id10597", "id10598", "id10599", "id10600", "id10601", "id10602", "id10603", "id10604", "id10605", "id10606", "id10607", "id10608", "id10609", "id10610", "id10611", "id10612", "id10613", "id10614", "id10615", "id10616", "id10617", "id10618", "id10619", "id10620", "id10621", "id10622", "id10623", "id10624", "id10625", "id10626", "id10627", "id10628", "id10629", "id10630", "id10631", "id10632", "id10633", "id10634", "id10635", "id10636", "id10637", "id10638", "id10639", "id10640", "id10641", "id10642", "id10643", "id10644", "id10645", "id10646", "id10647", "id10648", "id10649", "id10650", "id10651", "id10652", "id10653", "id10654", "id10655", "id10656", "id10657", "id10658", "id10659", "id10660", "id10661", "id10662", "id10663", "id10664", "id10665", "id10666", "id10667", "id10668", "id10669", "id10670", "id10671", "id10672", "id10673", "id10674", "id10675", "id10676", "id10677", "id10678", "id10679", "id10680", "id10681", "id10682", "id10683", "id10684", "id10685", "id10686", "id10687", "id10688", "id10689", "id10690", "id10691", "id10692", "id10693", "id10694", "id10695", "id10696", "id10697", "id10698", "id10699", "id10700", "id10701", "id10702", "id10703", "id10704", "id10705", "id10706", "id10707", "id10708", "id10709", "id10710", "id10711", "id10712", "id10713", "id10714", "id10715", "id10716", "id10717", "id10718", "id10719", "id10720", "id10721", "id10722", "id10723", "id10724", "id10725", "id10726", "id10727", "id10728", "id10729", "id10730", "id10731", "id10732", "id10733", "id10734", "id10735", "id10736", "id10737", "id10738", "id10739", "id10740", "id10741", "id10742", "id10743", "id10744", "id10745", "id10746", "id10747", "id10748", "id10749", "id10750", "id10751", "id10752", "id10753", "id10754", "id10755", "id10756", "id10757", "id10758", "id10759", "id10760", "id10761", "id10762", "id10763", "id10764", "id10765", "id10766", "id10767", "id10768", "id10769", "id10770", "id10771", "id10772", "id10773", "id10774", "id10775", "id10776", "id10777", "id10778", "id10779", "id10780", "id10781", "id10782", "id10783", "id10784", "id10785", "id10786", "id10787", "id10788", "id10789", "id10790", "id10791", "id10792", "id10793", "id10794", "id10795", "id10796", "id10797", "id10798", "id10799", "id10800", "id10801", "id10802", "id10803", "id10804", "id10805", "id10806", "id10807", "id10808", "id10809", "id10810", "id10811", "id10812", "id10813", "id10814", "id10815", "id10816", "id10817", "id10818", "id10819", "id10820", "id10821", "id10822", "id10823", "id10824", "id10825", "id10826", "id10827", "id10828", "id10829", "id10830", "id10831", "id10832", "id10833", "id10834", "id10835", "id10836", "id10837", "id10838", "id10839", "id10840", "id10841", "id10842", "id10843", "id10844", "id10845", "id10846", "id10847", "id10848", "id10849", "id10850", "id10851", "id10852", "id10853", "id10854", "id10855", "id10856", "id10857", "id10858", "id10859", "id10860", "id10861", "id10862", "id10863", "id10864", "id10865", "id10866", "id10867", "id10868", "id10869", "id10870", "id10871", "id10872", "id10873", "id10874", "id10875", "id10876", "id10877", "id10878", "id10879", "id10880", "id10881", "id10882", "id10883", "id10884", "id10885", "id10886", "id10887", "id10888", "id10889", "id10890", "id10891", "id10892", "id10893", "id10894", "id10895", "id10896", "id10897", "id10898", "id10899", "id10900", "id10901", "id10902", "id10903", "id10904", "id10905", "id10906", "id10907", "id10908", "id10909", "id10910", "id10911", "id10912", "id10913", "id10914", "id10915", "id10916", "id10917", "id10918", "id10919", "id10920", "id10921", "id10922", "id10923", "id10924", "id10925", "id10926", "id10927", "id10928", "id10929", "id10930", "id10931", "id10932", "id10933", "id10934", "id10935", "id10936", "id10937", "id10938", "id10939", "id10940", "id10941", "id10942", "id10943", "id10944", "id10945", "id10946", "id10947", "id10948", "id10949", "id10950", "id10951", "id10952", "id10953", "id10954", "id10955", "id10956", "id10957", "id10958", "id10959", "id10960", "id10961", "id10962", "id10963", "id10964", "id10965", "id10966", "id10967", "id10968", "id10969", "id10970", "id10971", "id10972", "id10973", "id10974", "id10975", "id10976", "id10977", "id10978", "id10979", "id10980", "id10981", "id10982", "id10983", "id10984", "id10985", "id10986", "id10987", "id10988", "id10989", "id10990", "id10991", "id10992", "id10993", "id10994", "id10995", "id10996", "id10997", "id10998", "id10999", "id11000", "id11001", "id11002", "id11003", "id11004", "id11005", "id11006", "id11007", "id11008", "id11009", "id11010", "id11011", "id11012", "id11013", "id11014", "id11015", "id11016", "id11017", "id11018", "id11019", "id11020", "id11021", "id11022", "id11023", "id11024", "id11025", "id11026", "id11027", "id11028", "id11029", "id11030", "id11031", "id11032", "id11033", "id11034", "id11035", "id11036", "id11037", "id11038", "id11039", "id11040", "id11041", "id11042", "id11043", "id11044", "id11045", "id11046", "id11047", "id11048", "id11049", "id11050", "id11051", "id11052", "id11053", "id11054", "id11055", "id11056", "id11057", "id11058", "id11059", "id11060", "id11061", "id11062", "id11063", "id11064", "id11065", "id11066", "id11067", "id11068", "id11069", "id11070", "id11071", "id11072", "id11073", "id11074", "id11075", "id11076", "id11077", "id11078", "id11079", "id11080", "id11081", "id11082", "id11083", "id11084", "id11085", "id11086", "id11087", "id11088", "id11089", "id11090", "id11091", "id11092", "id11093", "id11094", "id11095", "id11096", "id11097", "id11098", "id11099", "id11100", "id11101", "id11102", "id11103", "id11104", "id11105", "id11106", "id11107", "id11108", "id11109", "id11110", "id11111", "id11112", "id11113", "id11114", "id11115", "id11116", "id11117", "id11118", "id11119", "id11120", "id11121", "id11122", "id11123", "id11124", "id11125", "id11126", "id11127", "id11128", "id11129", "id11130", "id11131", "id11132", "id11133", "id11134", "id11135", "id11136", "id11137", "id11138", "id11139", "id11140", "id11141", "id11142", "id11143", "id11144", "id11145", "id11146", "id11147", "id11148", "id11149", "id11150", "id11151", "id11152", "id11153", "id11154", "id11155", "id11156", "id11157", "id11158", "id11159", "id11160", "id11161", "id11162", "id11163", "id11164", "id11165", "id11166", "id11167", "id11168", "id11169", "id11170", "id11171", "id11172", "id11173", "id11174", "id11175", "id11176", "id11177", "id11178", "id11179", "id11180", "id11181", "id11182", "id11183", "id11184", "id11185", "id11186", "id11187", "id11188", "id11189", "id11190", "id11191", "id11192", "id11193", "id11194", "id11195", "id11196", "id11197", "id11198", "id11199", "id11200", "id11201", "id11202", "id11203", "id11204", "id11205", "id11206", "id11207", "id11208", "id11209", "id11210", "id11211", "id11212", "id11213", "id11214", "id11215", "id11216", "id11217", "id11218", "id11219", "id11220", "id11221", "id11222", "id11223", "id11224", "id11225", "id11226", "id11227", "id11228", "id11229", "id11230", "id11231", "id11232", "id11233", "id11234", "id11235", "id11236", "id11237", "id11238", "id11239", "id11240", "id11241", "id11242", "id11243", "id11244", "id11245", "id11246", "id11247", "id11248", "id11249", "id11250", "id11251"], "names_file": null, "id": null, "_type": "ClassLabel"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "label"}, "task_templates": null, "builder_name": "superb", "config_name": "si", "version": {"version_str": "1.9.0", "description": "", "major": 1, "minor": 9, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 12729268, "num_examples": 138361, "dataset_name": "superb"}, "validation": {"name": "validation", "num_bytes": 635172, "num_examples": 6904, "dataset_name": "superb"}, "test": {"name": "test", "num_bytes": 759096, "num_examples": 8251, "dataset_name": "superb"}}, "download_checksums": {}, "download_size": 0, "post_processing_size": null, "dataset_size": 14123536, "size_in_bytes": 14123536}}
superb.py CHANGED
@@ -16,7 +16,7 @@
16
  # Lint as: python3
17
  """SUPERB: Speech processing Universal PERformance Benchmark."""
18
 
19
-
20
  import glob
21
  import os
22
  import textwrap
@@ -81,8 +81,8 @@ benchmark toolkit to fuel the research in representation learning and general
81
  speech processing.
82
 
83
  Note that in order to limit the required storage for preparing this dataset, the
84
- audio is stored in the .flac format and is not converted to a float32 array. To
85
- convert, the audio file to a float32 array, please make use of the `.map()`
86
  function as follows:
87
 
88
 
@@ -105,8 +105,8 @@ class SuperbConfig(datasets.BuilderConfig):
105
  def __init__(
106
  self,
107
  features,
108
- data_url,
109
  url,
 
110
  supervised_keys=None,
111
  task_templates=None,
112
  **kwargs,
@@ -154,7 +154,7 @@ class Superb(datasets.GeneratorBasedBuilder):
154
  """\
155
  Keyword Spotting (KS) detects preregistered keywords by classifying utterances into a predefined set of
156
  words. The task is usually performed on-device for the fast response time. Thus, accuracy, model size, and
157
- inference time are all crucial. SUPERB uses the widely used [Speech Commands dataset v1.0] for the task.
158
  The dataset consists of ten classes of keywords, a class for silence, and an unknown class to include the
159
  false positive. The evaluation metric is accuracy (ACC)"""
160
  ),
@@ -183,6 +183,65 @@ class Superb(datasets.GeneratorBasedBuilder):
183
  url="https://www.tensorflow.org/datasets/catalog/speech_commands",
184
  data_url="http://download.tensorflow.org/data/{filename}",
185
  ),
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
186
  SuperbConfig(
187
  name="sd",
188
  description=textwrap.dedent(
@@ -213,8 +272,62 @@ class Superb(datasets.GeneratorBasedBuilder):
213
  url="https://github.com/ftshijt/LibriMix",
214
  data_url="https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/{split}/{filename}",
215
  ),
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
216
  ]
217
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
218
  def _info(self):
219
  return datasets.DatasetInfo(
220
  description=_DESCRIPTION,
@@ -260,6 +373,34 @@ class Superb(datasets.GeneratorBasedBuilder):
260
  name=datasets.Split.TEST, gen_kwargs={"archive_path": archive_path["test"], "split": "test"}
261
  ),
262
  ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
263
  elif self.config.name == "sd":
264
  splits = ["train", "dev", "test"]
265
  _DL_URLS = {
@@ -276,11 +417,20 @@ class Superb(datasets.GeneratorBasedBuilder):
276
  )
277
  for split in splits
278
  ]
 
 
 
 
 
 
 
 
 
279
 
280
  def _generate_examples(self, archive_path, split=None):
281
  """Generate examples."""
282
  if self.config.name == "asr":
283
- transcripts_glob = os.path.join(archive_path, "LibriSpeech", "*/*/*/*.txt")
284
  key = 0
285
  for transcript_path in sorted(glob.glob(transcripts_glob)):
286
  transcript_dir_path = os.path.dirname(transcript_path)
@@ -311,6 +461,35 @@ class Superb(datasets.GeneratorBasedBuilder):
311
  else:
312
  label = "_unknown_"
313
  yield key, {"file": audio_file, "label": label}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
314
  elif self.config.name == "sd":
315
  data = SdData(archive_path)
316
  args = SdArgs()
@@ -338,6 +517,27 @@ class Superb(datasets.GeneratorBasedBuilder):
338
  "speakers": speakers,
339
  }
340
  key += 1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
341
 
342
 
343
  class SdData:
@@ -453,7 +653,7 @@ def _get_speakers(rec, data, args):
453
 
454
 
455
  def _split_ks_files(archive_path, split):
456
- audio_path = os.path.join(archive_path, "**/*.wav")
457
  audio_paths = glob.glob(audio_path)
458
  if split == "test":
459
  # use all available files for the test archive
 
16
  # Lint as: python3
17
  """SUPERB: Speech processing Universal PERformance Benchmark."""
18
 
19
+ import csv
20
  import glob
21
  import os
22
  import textwrap
 
81
  speech processing.
82
 
83
  Note that in order to limit the required storage for preparing this dataset, the
84
+ audio is stored in the .wav format and is not converted to a float32 array. To
85
+ convert the audio file to a float32 array, please make use of the `.map()`
86
  function as follows:
87
 
88
 
 
105
  def __init__(
106
  self,
107
  features,
 
108
  url,
109
+ data_url=None,
110
  supervised_keys=None,
111
  task_templates=None,
112
  **kwargs,
 
154
  """\
155
  Keyword Spotting (KS) detects preregistered keywords by classifying utterances into a predefined set of
156
  words. The task is usually performed on-device for the fast response time. Thus, accuracy, model size, and
157
+ inference time are all crucial. SUPERB uses the widely used Speech Commands dataset v1.0 for the task.
158
  The dataset consists of ten classes of keywords, a class for silence, and an unknown class to include the
159
  false positive. The evaluation metric is accuracy (ACC)"""
160
  ),
 
183
  url="https://www.tensorflow.org/datasets/catalog/speech_commands",
184
  data_url="http://download.tensorflow.org/data/{filename}",
185
  ),
186
+ SuperbConfig(
187
+ name="ic",
188
+ description=textwrap.dedent(
189
+ """\
190
+ Intent Classification (IC) classifies utterances into predefined classes to determine the intent of
191
+ speakers. SUPERB uses the Fluent Speech Commands dataset, where each utterance is tagged with three intent
192
+ labels: action, object, and location. The evaluation metric is accuracy (ACC)."""
193
+ ),
194
+ features=datasets.Features(
195
+ {
196
+ "file": datasets.Value("string"),
197
+ "speaker_id": datasets.Value("string"),
198
+ "text": datasets.Value("string"),
199
+ "action": datasets.ClassLabel(
200
+ names=["activate", "bring", "change language", "deactivate", "decrease", "increase"]
201
+ ),
202
+ "object": datasets.ClassLabel(
203
+ names=[
204
+ "Chinese",
205
+ "English",
206
+ "German",
207
+ "Korean",
208
+ "heat",
209
+ "juice",
210
+ "lamp",
211
+ "lights",
212
+ "music",
213
+ "newspaper",
214
+ "none",
215
+ "shoes",
216
+ "socks",
217
+ "volume",
218
+ ]
219
+ ),
220
+ "location": datasets.ClassLabel(names=["bedroom", "kitchen", "none", "washroom"]),
221
+ }
222
+ ),
223
+ supervised_keys=None,
224
+ url="https://fluent.ai/fluent-speech-commands-a-dataset-for-spoken-language-understanding-research/",
225
+ data_url="http://fluent.ai:2052/jf8398hf30f0381738rucj3828chfdnchs.tar.gz",
226
+ ),
227
+ SuperbConfig(
228
+ name="si",
229
+ description=textwrap.dedent(
230
+ """\
231
+ Speaker Identification (SI) classifies each utterance for its speaker identity as a multi-class
232
+ classification, where speakers are in the same predefined set for both training and testing. The widely
233
+ used VoxCeleb1 dataset is adopted, and the evaluation metric is accuracy (ACC)."""
234
+ ),
235
+ features=datasets.Features(
236
+ {
237
+ "file": datasets.Value("string"),
238
+ # VoxCeleb1 contains 1251 speaker IDs in range ["id10001",..."id11251"]
239
+ "label": datasets.ClassLabel(names=[f"id{i + 10001}" for i in range(1251)]),
240
+ }
241
+ ),
242
+ supervised_keys=("file", "label"),
243
+ url="https://www.robots.ox.ac.uk/~vgg/data/voxceleb/vox1.html",
244
+ ),
245
  SuperbConfig(
246
  name="sd",
247
  description=textwrap.dedent(
 
272
  url="https://github.com/ftshijt/LibriMix",
273
  data_url="https://huggingface.co/datasets/superb/superb-data/resolve/main/sd/{split}/{filename}",
274
  ),
275
+ SuperbConfig(
276
+ name="er",
277
+ description=textwrap.dedent(
278
+ """\
279
+ Emotion Recognition (ER) predicts an emotion class for each utterance. The most widely used ER dataset
280
+ IEMOCAP is adopted, and we follow the conventional evaluation protocol: we drop the unbalanced emotion
281
+ classes to leave the final four classes with a similar amount of data points and cross-validate on five
282
+ folds of the standard splits. The evaluation metric is accuracy (ACC)."""
283
+ ),
284
+ features=datasets.Features(
285
+ {
286
+ "file": datasets.Value("string"),
287
+ "label": datasets.ClassLabel(names=["neu", "hap", "ang", "sad"]),
288
+ }
289
+ ),
290
+ supervised_keys=("file", "label"),
291
+ url="https://sail.usc.edu/iemocap/",
292
+ ),
293
  ]
294
 
295
+ @property
296
+ def manual_download_instructions(self):
297
+ if self.config.name == "si":
298
+ return textwrap.dedent(
299
+ """
300
+ Please download the VoxCeleb dataset using the following script,
301
+ which should create `VoxCeleb1/wav/id*` directories for both train and test speakers`:
302
+ ```
303
+ mkdir VoxCeleb1
304
+ cd VoxCeleb1
305
+
306
+ wget https://thor.robots.ox.ac.uk/~vgg/data/voxceleb/vox1a/vox1_dev_wav_partaa
307
+ wget https://thor.robots.ox.ac.uk/~vgg/data/voxceleb/vox1a/vox1_dev_wav_partab
308
+ wget https://thor.robots.ox.ac.uk/~vgg/data/voxceleb/vox1a/vox1_dev_wav_partac
309
+ wget https://thor.robots.ox.ac.uk/~vgg/data/voxceleb/vox1a/vox1_dev_wav_partad
310
+ cat vox1_dev* > vox1_dev_wav.zip
311
+ unzip vox1_dev_wav.zip
312
+
313
+ wget https://thor.robots.ox.ac.uk/~vgg/data/voxceleb/vox1a/vox1_test_wav.zip
314
+ unzip vox1_test_wav.zip
315
+
316
+ # download the official SUPERB train-dev-test split
317
+ wget https://raw.githubusercontent.com/s3prl/s3prl/master/s3prl/downstream/voxceleb1/veri_test_class.txt
318
+ ```"""
319
+ )
320
+ elif self.config.name == "er":
321
+ return textwrap.dedent(
322
+ """
323
+ Please download the IEMOCAP dataset after submitting the request form here:
324
+ https://sail.usc.edu/iemocap/iemocap_release.htm
325
+ Having downloaded the dataset you can extract it with `tar -xvzf IEMOCAP_full_release.tar.gz`
326
+ which should create a folder called `IEMOCAP_full_release`
327
+ """
328
+ )
329
+ return None
330
+
331
  def _info(self):
332
  return datasets.DatasetInfo(
333
  description=_DESCRIPTION,
 
373
  name=datasets.Split.TEST, gen_kwargs={"archive_path": archive_path["test"], "split": "test"}
374
  ),
375
  ]
376
+ elif self.config.name == "ic":
377
+ archive_path = dl_manager.download_and_extract(self.config.data_url)
378
+ return [
379
+ datasets.SplitGenerator(
380
+ name=datasets.Split.TRAIN,
381
+ gen_kwargs={"archive_path": archive_path, "split": "train"},
382
+ ),
383
+ datasets.SplitGenerator(
384
+ name=datasets.Split.VALIDATION,
385
+ gen_kwargs={"archive_path": archive_path, "split": "valid"},
386
+ ),
387
+ datasets.SplitGenerator(
388
+ name=datasets.Split.TEST, gen_kwargs={"archive_path": archive_path, "split": "test"}
389
+ ),
390
+ ]
391
+ elif self.config.name == "si":
392
+ manual_dir = os.path.abspath(os.path.expanduser(dl_manager.manual_dir))
393
+ return [
394
+ datasets.SplitGenerator(
395
+ name=datasets.Split.TRAIN,
396
+ gen_kwargs={"archive_path": manual_dir, "split": 1},
397
+ ),
398
+ datasets.SplitGenerator(
399
+ name=datasets.Split.VALIDATION,
400
+ gen_kwargs={"archive_path": manual_dir, "split": 2},
401
+ ),
402
+ datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"archive_path": manual_dir, "split": 3}),
403
+ ]
404
  elif self.config.name == "sd":
405
  splits = ["train", "dev", "test"]
406
  _DL_URLS = {
 
417
  )
418
  for split in splits
419
  ]
420
+ elif self.config.name == "er":
421
+ manual_dir = os.path.abspath(os.path.expanduser(dl_manager.manual_dir))
422
+ return [
423
+ datasets.SplitGenerator(
424
+ name=f"session{i}",
425
+ gen_kwargs={"archive_path": manual_dir, "split": i},
426
+ )
427
+ for i in range(1, 6)
428
+ ]
429
 
430
  def _generate_examples(self, archive_path, split=None):
431
  """Generate examples."""
432
  if self.config.name == "asr":
433
+ transcripts_glob = os.path.join(archive_path, "LibriSpeech", "*", "*", "*", "*.txt")
434
  key = 0
435
  for transcript_path in sorted(glob.glob(transcripts_glob)):
436
  transcript_dir_path = os.path.dirname(transcript_path)
 
461
  else:
462
  label = "_unknown_"
463
  yield key, {"file": audio_file, "label": label}
464
+ elif self.config.name == "ic":
465
+ root_path = os.path.join(archive_path, "fluent_speech_commands_dataset")
466
+ csv_path = os.path.join(root_path, "data", f"{split}_data.csv")
467
+ with open(csv_path, encoding="utf-8") as csv_file:
468
+ csv_reader = csv.reader(csv_file, delimiter=",", skipinitialspace=True)
469
+ next(csv_reader)
470
+ for row in csv_reader:
471
+ key, file_path, speaker_id, text, action, object_, location = row
472
+ yield key, {
473
+ "file": os.path.join(root_path, file_path),
474
+ "speaker_id": speaker_id,
475
+ "text": text,
476
+ "action": action,
477
+ "object": object_,
478
+ "location": location,
479
+ }
480
+ elif self.config.name == "si":
481
+ wav_path = os.path.join(archive_path, "wav")
482
+ splits_path = os.path.join(archive_path, "veri_test_class.txt")
483
+ with open(splits_path, "r", encoding="utf-8") as f:
484
+ for key, line in enumerate(f):
485
+ split_id, file_path = line.strip().split(" ")
486
+ if int(split_id) != split:
487
+ continue
488
+ speaker_id = file_path.split("/")[0]
489
+ yield key, {
490
+ "file": os.path.join(wav_path, file_path),
491
+ "label": speaker_id,
492
+ }
493
  elif self.config.name == "sd":
494
  data = SdData(archive_path)
495
  args = SdArgs()
 
517
  "speakers": speakers,
518
  }
519
  key += 1
520
+ elif self.config.name == "er":
521
+ root_path = os.path.join(archive_path, f"Session{split}")
522
+ wav_path = os.path.join(root_path, "sentences", "wav")
523
+ labels_path = os.path.join(root_path, "dialog", "EmoEvaluation", "*.txt")
524
+ emotions = ["neu", "hap", "ang", "sad", "exc"]
525
+ key = 0
526
+ for labels_file in sorted(glob.glob(labels_path)):
527
+ with open(labels_file, "r", encoding="utf-8") as f:
528
+ for line in f:
529
+ if line[0] != "[":
530
+ continue
531
+ _, filename, emo, _ = line.split("\t")
532
+ if emo not in emotions:
533
+ continue
534
+ wav_subdir = filename.rsplit("_", 1)[0]
535
+ filename = f"{filename}.wav"
536
+ yield key, {
537
+ "file": os.path.join(wav_path, wav_subdir, filename),
538
+ "label": emo.replace("exc", "hap"),
539
+ }
540
+ key += 1
541
 
542
 
543
  class SdData:
 
653
 
654
 
655
  def _split_ks_files(archive_path, split):
656
+ audio_path = os.path.join(archive_path, "**", "*.wav")
657
  audio_paths = glob.glob(audio_path)
658
  if split == "test":
659
  # use all available files for the test archive