support streaming

README.md

@@ -79,9 +79,9 @@ configs:
 - **Repository:** [https://github.com/asappresearch/slue-toolkit/](https://github.com/asappresearch/slue-toolkit/)
 - **Paper:** [https://arxiv.org/pdf/2111.10367.pdf](https://arxiv.org/pdf/2111.10367.pdf)
 - **Leaderboard:** [https://asappresearch.github.io/slue-toolkit/leaderboard_v0.2.html](https://asappresearch.github.io/slue-toolkit/leaderboard_v0.2.html)
-- **Size of downloaded dataset files:** 1.93 GB
+- **Size of downloaded dataset files:** 1.95 GB
 - **Size of the generated dataset:** 9.59 MB
-- **Total amount of disk used:** 1.94 GB
+- **Total amount of disk used:** 1.95 GB
 
 ### Dataset Summary
 
@@ -119,9 +119,9 @@ The language data in SLUE is in English.
 ### Data Instances
 
 #### voxpopuli
-- **Size of downloaded dataset files:** 389.31 MB
+- **Size of downloaded dataset files:** 398.45 MB
 - **Size of the generated dataset:** 5.81 MB
-- **Total amount of disk used:** 395.12 MB
+- **Total amount of disk used:** 404.26 MB
 An example of 'train' looks as follows.
 ```
 {'id': '20131007-0900-PLENARY-19-en_20131007-21:26:04_3',

dataset_infos.json

@@ -1 +1 @@
-{"voxpopuli": {"description": "Spoken Language Understanding Evaluation (SLUE) benchmark. There are two subsets: (i) SLUE-VoxPopuli which has ASR and NER tasks and (ii) SLUE-VoxCeleb which has ASR and SA tasks.\n", "citation": "@inproceedings{shon2022slue,\n  title={Slue: New benchmark tasks for spoken language understanding evaluation on natural speech},\n  author={Shon, Suwon and Pasad, Ankita and Wu, Felix and Brusco, Pablo and Artzi, Yoav and Livescu, Karen and Han, Kyu J},\n  booktitle={ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n  pages={7927--7931},\n  year={2022},\n  organization={IEEE}\n}\n", "homepage": "https://asappresearch.github.io/slue-toolkit/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "audio": {"sampling_rate": 16000, "mono": true, "decode": true, "id": null, "_type": "Audio"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "normalized_text": {"dtype": "string", "id": null, "_type": "Value"}, "raw_text": {"dtype": "string", "id": null, "_type": "Value"}, "raw_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "normalized_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "raw_combined_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "normalized_combined_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "text"}, "task_templates": null, "builder_name": "slue", "config_name": "voxpopuli", "version": {"version_str": "2.4.0", "description": "", "major": 2, "minor": 4, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3527760, "num_examples": 5000, "dataset_name": "slue"}, "validation": {"name": "validation", "num_bytes": 1181375, "num_examples": 1753, "dataset_name": "slue"}, "test": {"name": "test", "num_bytes": 1100635, "num_examples": 1842, "dataset_name": "slue"}}, "download_checksums": {"https://public-dataset-model-store.awsdev.asapp.com/users/sshon/public/slue/slue-voxpopuli_v0.2_blind.tar.gz": {"num_bytes": 389309633, "checksum": "e7cf96d3d98914a70c646e1958eb18e3d4fbf7e40fafa9f4cc3e814bcbc2924f"}}, "download_size": 389309633, "post_processing_size": null, "dataset_size": 5809770, "size_in_bytes": 395119403}, "voxceleb": {"description": "Spoken Language Understanding Evaluation (SLUE) benchmark. There are two subsets: (i) SLUE-VoxPopuli which has ASR and NER tasks and (ii) SLUE-VoxCeleb which has ASR and SA tasks.\n", "citation": "@inproceedings{shon2022slue,\n  title={Slue: New benchmark tasks for spoken language understanding evaluation on natural speech},\n  author={Shon, Suwon and Pasad, Ankita and Wu, Felix and Brusco, Pablo and Artzi, Yoav and Livescu, Karen and Han, Kyu J},\n  booktitle={ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n  pages={7927--7931},\n  year={2022},\n  organization={IEEE}\n}\n", "homepage": "https://asappresearch.github.io/slue-toolkit/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "audio": {"sampling_rate": 16000, "mono": true, "decode": true, "id": null, "_type": "Audio"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "normalized_text": {"dtype": "string", "id": null, "_type": "Value"}, "sentiment": {"dtype": "string", "id": null, "_type": "Value"}, "start_second": {"dtype": "float64", "id": null, "_type": "Value"}, "end_second": {"dtype": "float64", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "text"}, "task_templates": null, "builder_name": "slue", "config_name": "voxceleb", "version": {"version_str": "2.4.0", "description": "", "major": 2, "minor": 4, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2311287, "num_examples": 5777, "dataset_name": "slue"}, "validation": {"name": "validation", "num_bytes": 566001, "num_examples": 1454, "dataset_name": "slue"}, "test": {"name": "test", "num_bytes": 899354, "num_examples": 3553, "dataset_name": "slue"}}, "download_checksums": {"https://public-dataset-model-store.awsdev.asapp.com/users/sshon/public/slue/slue-voxceleb_v0.2_blind.tar.gz": {"num_bytes": 1544292490, "checksum": "6564fa24232b8db9f994a0a3dc6100db4b9dd5a804062c9b70cab51da25eb193"}}, "download_size": 1544292490, "post_processing_size": null, "dataset_size": 3776642, "size_in_bytes": 1548069132}}
+{"voxpopuli": {"description": "Spoken Language Understanding Evaluation (SLUE) benchmark. There are two subsets: (i) SLUE-VoxPopuli which has ASR and NER tasks and (ii) SLUE-VoxCeleb which has ASR and SA tasks.\n", "citation": "@inproceedings{shon2022slue,\n  title={Slue: New benchmark tasks for spoken language understanding evaluation on natural speech},\n  author={Shon, Suwon and Pasad, Ankita and Wu, Felix and Brusco, Pablo and Artzi, Yoav and Livescu, Karen and Han, Kyu J},\n  booktitle={ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n  pages={7927--7931},\n  year={2022},\n  organization={IEEE}\n}\n", "homepage": "https://asappresearch.github.io/slue-toolkit/", "license": "\n=======================================================\nThe license of this script\n\nMIT License\n\nCopyright (c) 2022 ASAPP Inc.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.\n\n=======================================================\nSLUE-VoxPopuli Dataset\n\nSLUE-VoxPopuli dataset contains a subset of VoxPopuli dataset and the copyright of this subset remains the same with the original license, CC0. See also European Parliament's legal notice (https://www.europarl.europa.eu/legal-notice/en/)\n\nAdditionally, we provide named entity annotation (normalized_ner and raw_ner column in .tsv files) and it is covered with the same license as CC0.\n=======================================================\nSLUE-VoxCeleb Dataset\n\nSLUE-VoxCeleb Dataset contains a subset of OXFORD VoxCeleb dataset and the copyright of this subset remains the same Creative Commons Attribution 4.0 International license as below. Additionally, we provide transcription, sentiment annotation and timestamp (start, end) that follows the same license to OXFORD VoxCeleb dataset.\n\n=======================================================\nOXFORD VGG VoxCeleb Dataset \n\nVoxCeleb1 contains over 100,000 utterances for 1,251 celebrities, extracted from videos uploaded to YouTube.  \nVoxCeleb2 contains over a million utterances for 6,112 celebrities, extracted from videos uploaded to YouTube. \n\nThe speakers span a wide range of different ethnicities, accents, professions and ages. \n\nWe provide Youtube URLs, associated face detections, and timestamps, as\nwell as cropped audio segments and cropped face videos from the\ndataset.  The copyright of both the original and cropped versions\nof the videos remains with the original owners.\n\nThe data is covered under a Creative Commons\nAttribution 4.0 International license (Please read the\nlicense terms here. https://creativecommons.org/licenses/by/4.0/).\n\nDownloading this dataset implies agreement to follow the same\nconditions for any modification and/or\nre-distribution of the dataset in any form.\n\n\nAdditionally any entity using this dataset agrees to the following conditions:\n\nTHIS DATASET IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS \"AS\nIS\" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED\nTO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A\nPARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT\nHOLDER BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,\nEXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,\nPROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR\nPROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF\nLIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING\nNEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS\nSOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.\n\n\nPlease cite [1,2] below if you make use of the dataset.\n\n[1] J. S. Chung, A. Nagrani, A. Zisserman  \nVoxCeleb2: Deep Speaker Recognition  \nINTERSPEECH, 2018.\n\n[2] A. Nagrani, J. S. Chung, A. Zisserman\nVoxCeleb: a large-scale speaker identification dataset  \nINTERSPEECH, 2017\n=======================================================\n\n", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "audio": {"sampling_rate": 16000, "mono": true, "decode": true, "id": null, "_type": "Audio"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "normalized_text": {"dtype": "string", "id": null, "_type": "Value"}, "raw_text": {"dtype": "string", "id": null, "_type": "Value"}, "raw_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "normalized_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "raw_combined_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "normalized_combined_ner": {"feature": {"type": {"dtype": "string", "id": null, "_type": "Value"}, "start": {"dtype": "int32", "id": null, "_type": "Value"}, "length": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "text"}, "task_templates": null, "builder_name": "slue", "config_name": "voxpopuli", "version": {"version_str": "2.4.0", "description": "", "major": 2, "minor": 4, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3527760, "num_examples": 5000, "dataset_name": "slue"}, "validation": {"name": "validation", "num_bytes": 1181375, "num_examples": 1753, "dataset_name": "slue"}, "test": {"name": "test", "num_bytes": 1100635, "num_examples": 1842, "dataset_name": "slue"}}, "download_checksums": {"https://public-dataset-model-store.awsdev.asapp.com/users/sshon/public/slue/slue-voxpopuli_v0.2_blind.zip": {"num_bytes": 398450006, "checksum": "21d12d2bcb44f8dd78885211f381f4443c9a18ea5703b21d5f43460f0661c53e"}}, "download_size": 398450006, "post_processing_size": null, "dataset_size": 5809770, "size_in_bytes": 404259776}, "voxceleb": {"description": "Spoken Language Understanding Evaluation (SLUE) benchmark. There are two subsets: (i) SLUE-VoxPopuli which has ASR and NER tasks and (ii) SLUE-VoxCeleb which has ASR and SA tasks.\n", "citation": "@inproceedings{shon2022slue,\n  title={Slue: New benchmark tasks for spoken language understanding evaluation on natural speech},\n  author={Shon, Suwon and Pasad, Ankita and Wu, Felix and Brusco, Pablo and Artzi, Yoav and Livescu, Karen and Han, Kyu J},\n  booktitle={ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n  pages={7927--7931},\n  year={2022},\n  organization={IEEE}\n}\n", "homepage": "https://asappresearch.github.io/slue-toolkit/", "license": "\n=======================================================\nThe license of this script\n\nMIT License\n\nCopyright (c) 2022 ASAPP Inc.\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.\n\n=======================================================\nSLUE-VoxPopuli Dataset\n\nSLUE-VoxPopuli dataset contains a subset of VoxPopuli dataset and the copyright of this subset remains the same with the original license, CC0. See also European Parliament's legal notice (https://www.europarl.europa.eu/legal-notice/en/)\n\nAdditionally, we provide named entity annotation (normalized_ner and raw_ner column in .tsv files) and it is covered with the same license as CC0.\n=======================================================\nSLUE-VoxCeleb Dataset\n\nSLUE-VoxCeleb Dataset contains a subset of OXFORD VoxCeleb dataset and the copyright of this subset remains the same Creative Commons Attribution 4.0 International license as below. Additionally, we provide transcription, sentiment annotation and timestamp (start, end) that follows the same license to OXFORD VoxCeleb dataset.\n\n=======================================================\nOXFORD VGG VoxCeleb Dataset \n\nVoxCeleb1 contains over 100,000 utterances for 1,251 celebrities, extracted from videos uploaded to YouTube.  \nVoxCeleb2 contains over a million utterances for 6,112 celebrities, extracted from videos uploaded to YouTube. \n\nThe speakers span a wide range of different ethnicities, accents, professions and ages. \n\nWe provide Youtube URLs, associated face detections, and timestamps, as\nwell as cropped audio segments and cropped face videos from the\ndataset.  The copyright of both the original and cropped versions\nof the videos remains with the original owners.\n\nThe data is covered under a Creative Commons\nAttribution 4.0 International license (Please read the\nlicense terms here. https://creativecommons.org/licenses/by/4.0/).\n\nDownloading this dataset implies agreement to follow the same\nconditions for any modification and/or\nre-distribution of the dataset in any form.\n\n\nAdditionally any entity using this dataset agrees to the following conditions:\n\nTHIS DATASET IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS \"AS\nIS\" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED\nTO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A\nPARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT\nHOLDER BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,\nEXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,\nPROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR\nPROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF\nLIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING\nNEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS\nSOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.\n\n\nPlease cite [1,2] below if you make use of the dataset.\n\n[1] J. S. Chung, A. Nagrani, A. Zisserman  \nVoxCeleb2: Deep Speaker Recognition  \nINTERSPEECH, 2018.\n\n[2] A. Nagrani, J. S. Chung, A. Zisserman\nVoxCeleb: a large-scale speaker identification dataset  \nINTERSPEECH, 2017\n=======================================================\n\n", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "audio": {"sampling_rate": 16000, "mono": true, "decode": true, "id": null, "_type": "Audio"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "normalized_text": {"dtype": "string", "id": null, "_type": "Value"}, "sentiment": {"dtype": "string", "id": null, "_type": "Value"}, "start_second": {"dtype": "float64", "id": null, "_type": "Value"}, "end_second": {"dtype": "float64", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "file", "output": "text"}, "task_templates": null, "builder_name": "slue", "config_name": "voxceleb", "version": {"version_str": "2.4.0", "description": "", "major": 2, "minor": 4, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2311287, "num_examples": 5777, "dataset_name": "slue"}, "validation": {"name": "validation", "num_bytes": 566001, "num_examples": 1454, "dataset_name": "slue"}, "test": {"name": "test", "num_bytes": 899354, "num_examples": 3553, "dataset_name": "slue"}}, "download_checksums": {"https://public-dataset-model-store.awsdev.asapp.com/users/sshon/public/slue/slue-voxceleb_v0.2_blind.zip": {"num_bytes": 1545458156, "checksum": "74b3bfa2d6555cab10821be45c2289318e87efc698ef043eb231f2a738a91af3"}}, "download_size": 1545458156, "post_processing_size": null, "dataset_size": 3776642, "size_in_bytes": 1549234798}}

slue.py

@@ -1,5 +1,26 @@
-"""
+from typing import List
+import os
+import csv
+import ast
+import gzip
+
+import datasets
+from datasets.utils.logging import get_logger
+
+logger = get_logger(__name__)
+
+_URL = "https://asappresearch.github.io/slue-toolkit/"
+_DL_URL = "https://public-dataset-model-store.awsdev.asapp.com/users/sshon/public/slue/"
+
+_DL_URLS = {
+    "slue-voxpopuli": _DL_URL + "slue-voxpopuli_v0.2_blind.zip",
+    "slue-voxceleb": _DL_URL + "slue-voxceleb_v0.2_blind.zip",
+}
+
+_LICENSE = """
 =======================================================
+The license of this script
+
 MIT License
 
 Copyright (c) 2022 ASAPP Inc.
@@ -83,24 +104,6 @@ INTERSPEECH, 2017
 
 """
 
-from typing import List
-import os
-import csv
-import ast
-
-import datasets
-from datasets.utils.logging import get_logger
-
-logger = get_logger(__name__)
-
-_URL = "https://asappresearch.github.io/slue-toolkit/"
-_DL_URL = "https://public-dataset-model-store.awsdev.asapp.com/users/sshon/public/slue/"
-
-_DL_URLS = {
-    "slue-voxpopuli": _DL_URL + "slue-voxpopuli_v0.2_blind.tar.gz",
-    "slue-voxceleb": _DL_URL + "slue-voxceleb_v0.2_blind.tar.gz",
-}
-
 _CITATION = """\
 @inproceedings{shon2022slue,
   title={Slue: New benchmark tasks for spoken language understanding evaluation on natural speech},
@@ -240,6 +243,7 @@ class SLUE(datasets.GeneratorBasedBuilder):
             supervised_keys=("file", "text"),
             homepage=_URL,
             citation=_CITATION,
+            license=_LICENSE,
         )
 
     def _split_generators(