{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/slue-new-benchmark-tasks-for-spoken-language","title":"SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech","arxiv_id":"2111.10367","date":"2021-11-19","proceeding":null,"authors":["Suwon Shon","Ankita Pasad","Felix Wu","Pablo Brusco","Yoav Artzi","Karen Livescu","Kyu J. Han"],"abstract":"Progress in speech processing has been facilitated by shared datasets and benchmarks. Historically these have focused on automatic speech recognition (ASR), speaker identification, or other lower-level tasks. Interest has been growing in higher-level spoken language understanding tasks, including using end-to-end models, but there are fewer annotated datasets for such tasks. At the same time, recent work shows the possibility of pre-training generic representations and then fine-tuning for several tasks using relatively little labeled data. We propose to create a suite of benchmark tasks for Spoken Language Understanding Evaluation (SLUE) consisting of limited-size labeled training sets and corresponding evaluation sets. This resource would allow the research community to track progress, evaluate pre-trained representations for higher-level tasks, and study open questions such as the utility of pipeline versus end-to-end approaches. We present the first phase of the SLUE benchmark suite, consisting of named entity recognition, sentiment analysis, and ASR on the corresponding datasets. We focus on naturally produced (not read or synthesized) speech, and freely available datasets. We provide new transcriptions and annotations on subsets of the VoxCeleb and VoxPopuli datasets, evaluation metrics and results for baseline models, and an open-source toolkit to reproduce the baselines and evaluate new models.","url_abs":"https://arxiv.org/abs/2111.10367v3","url_pdf":"https://arxiv.org/pdf/2111.10367v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"slue-new-benchmark-tasks-for-spoken-language","repo_url":"https://github.com/asappresearch/slue-toolkit","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"automatic-speech-recognition-2","task_name":"Automatic Speech Recognition"},{"task_slug":"automatic-speech-recognition","task_name":"Automatic Speech Recognition (ASR)"},{"task_slug":"named-entity-recognition-1","task_name":"Named Entity Recognition"},{"task_slug":"named-entity-recognition-ner","task_name":"Named Entity Recognition (NER)"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"speaker-identification","task_name":"Speaker Identification"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"spoken-language-understanding","task_name":"Spoken Language Understanding"},{"task_slug":"named-entity-recognition","task_name":"named-entity-recognition"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[],"datasets_introduced":[{"slug":"slue","name":"SLUE","full_name":"Spoken Language Understanding Evaluation"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-L-LL60K (pipeline approach, uses LM)","rank_in_archive_order":1,"of":13,"metrics":{"F1 (%)":"69.6","Text model":"DeBERTa-L","label-F1 (%)":"82.2"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-B-LS960 (pipeline approach, uses LM)","rank_in_archive_order":2,"of":13,"metrics":{"F1 (%)":"68.0","Text model":"DeBERTa-L","label-F1 (%)":"79.8"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-L-LL60K (e2e approach, uses LM)","rank_in_archive_order":4,"of":13,"metrics":{"F1 (%)":"64.8","Text model":"N/A","label-F1 (%)":"73.3"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-B-LS960 (e2e approach, uses LM)","rank_in_archive_order":5,"of":13,"metrics":{"F1 (%)":"63.4","Text model":"N/A","label-F1 (%)":"71.7"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"HuBERT-B-LS960 (e2e approach, uses LM)","rank_in_archive_order":6,"of":13,"metrics":{"F1 (%)":"61.9","Text model":"N/A","label-F1 (%)":"70.3"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-B-VP100K (e2e approach, uses LM)","rank_in_archive_order":7,"of":13,"metrics":{"F1 (%)":"61.8","Text model":"N/A","label-F1 (%)":"69.8"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-L-LL60K (pipeline approach)","rank_in_archive_order":8,"of":13,"metrics":{"F1 (%)":"57.8","Text model":"DeBERTa-L","label-F1 (%)":"78.8"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-L-LL60K (e2e approach)","rank_in_archive_order":9,"of":13,"metrics":{"F1 (%)":"50.9","Text model":"-","label-F1 (%)":"64.7"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-B-LS960 (e2e approach)","rank_in_archive_order":10,"of":13,"metrics":{"F1 (%)":"50.2","Text model":"-","label-F1 (%)":"64.0"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"HuBERT-B-LS960 (e2e approach)","rank_in_archive_order":11,"of":13,"metrics":{"F1 (%)":"49.8","Text model":"-","label-F1 (%)":"62.9"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-B-LS960 (pipeline approach)","rank_in_archive_order":12,"of":13,"metrics":{"F1 (%)":"49.5","Text model":"DeBERTa-L","label-F1 (%)":"74.2"},"uses_additional_data":false},{"leaderboard":"/sota/named-entity-recognition-on-slue","task":"Named Entity Recognition (NER)","dataset":"SLUE","model":"W2V2-B-VP100K (e2e approach)","rank_in_archive_order":13,"of":13,"metrics":{"F1 (%)":"47.9","Text model":"-","label-F1 (%)":"60.8"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-L-LL60K (pipeline approach, uses LM)","rank_in_archive_order":1,"of":8,"metrics":{"F1 (%)":"63.3","Recall (%)\t":"60.4","Text model":"DeBERTa-L"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-L-LL60K (pipeline approach)","rank_in_archive_order":2,"of":8,"metrics":{"F1 (%)":"63.3","Recall (%)\t":"60.2","Text model":"DeBERTa-L"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-B-LS960 (pipeline approach, uses LM)","rank_in_archive_order":3,"of":8,"metrics":{"F1 (%)":"62.9","Recall (%)\t":"60.0","Text model":"DeBERTa-L"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-B-LS960 (pipeline approach)","rank_in_archive_order":4,"of":8,"metrics":{"F1 (%)":"61.8","Recall (%)\t":"59.0","Text model":"DeBERTa-L"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-L-LL60K (e2e approach)","rank_in_archive_order":5,"of":8,"metrics":{"F1 (%)":"48.5","Recall (%)\t":"49.2","Text model":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"HuBERT-B-LS960 (e2e approach)","rank_in_archive_order":6,"of":8,"metrics":{"F1 (%)":"48.0","Recall (%)\t":"47.5","Text model":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-B-LS960 (e2e approach)","rank_in_archive_order":7,"of":8,"metrics":{"F1 (%)":"46.6","Recall (%)\t":"46.0","Text model":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-slue","task":"Sentiment Analysis","dataset":"SLUE","model":"W2V2-B-VP100K (e2e approach)","rank_in_archive_order":8,"of":8,"metrics":{"F1 (%)":"38.4","Recall (%)\t":"38.7","Text model":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-L-LL60K (+ TED-LIUM 3 LM)","rank_in_archive_order":1,"of":8,"metrics":{"VoxCeleb (Dev)":"9.1","VoxCeleb (Test)":"10.8","VoxPopuli (Dev)":"9.1","VoxPopuli (Test)":"9.3"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-B-LS960 (+ TED-LIUM 3 LM)","rank_in_archive_order":2,"of":8,"metrics":{"VoxCeleb (Dev)":"13.2","VoxCeleb (Test)":"15.8","VoxPopuli (Dev)":"12.0","VoxPopuli (Test)":"12.2"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-L-LL60K (+ in-domain LM)","rank_in_archive_order":3,"of":8,"metrics":{"VoxCeleb (Dev)":"11.8","VoxCeleb (Test)":"13.8","VoxPopuli (Dev)":"12.0","VoxPopuli (Test)":"12.5"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-L-LL60K","rank_in_archive_order":4,"of":8,"metrics":{"VoxCeleb (Dev)":"11.0","VoxCeleb (Test)":"13.5","VoxPopuli (Dev)":"14.0","VoxPopuli (Test)":"12.1"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-B-LS960 (+ in-domain LM)","rank_in_archive_order":5,"of":8,"metrics":{"VoxCeleb (Dev)":"15.2","VoxCeleb (Test)":"18.2","VoxPopuli (Dev)":"14.6","VoxPopuli (Test)":"15.2"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-B-LS960","rank_in_archive_order":6,"of":8,"metrics":{"VoxCeleb (Dev)":"17.2","VoxCeleb (Test)":"20.5","VoxPopuli (Dev)":"17.2","VoxPopuli (Test)":"17.9"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"HuBERT-B-LS960","rank_in_archive_order":7,"of":8,"metrics":{"VoxCeleb (Dev)":"19.6","VoxCeleb (Test)":"21.2","VoxPopuli (Dev)":"18.6","VoxPopuli (Test)":"19.1"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-slue","task":"Speech Recognition","dataset":"SLUE","model":"W2V2-B-VP100K","rank_in_archive_order":8,"of":8,"metrics":{"VoxCeleb (Dev)":"29.9","VoxCeleb (Test)":"33.4","VoxPopuli (Dev)":"21.6","VoxPopuli (Test)":"22.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2111.10367","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2111.10367"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/asappresearch/slue-toolkit","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":12},"by_repo_kind":{"official":{"samples":12,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"0302aff920740fbc","entry":"convert","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_utils_nel.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_utils_nel.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0302aff920740fbc"}},{"code_sha256_prefix":"3b9788e17612ec5c","entry":"eval_ctc_model","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_w2v.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_w2v.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3b9788e17612ec5c"}},{"code_sha256_prefix":"c8831cc7145df3ab","entry":"evaluate","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_utils_nel.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_utils_nel.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c8831cc7145df3ab"}},{"code_sha256_prefix":"7df2656f9761c553","entry":"get_ner_scores","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_utils_ner.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_utils_ner.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7df2656f9761c553"}},{"code_sha256_prefix":"7f746dc6c2a62bb2","entry":"get_ner_stats","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_utils_ner.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_utils_ner.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7f746dc6c2a62bb2"}},{"code_sha256_prefix":"656dc5c1fcd38ff8","entry":"load_dct","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/generic_utils.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/generic_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"656dc5c1fcd38ff8"}},{"code_sha256_prefix":"16afefff56a55570","entry":"load_json","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/generic_utils.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/generic_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"16afefff56a55570"}},{"code_sha256_prefix":"e29d3eb7f31a6d05","entry":"load_pkl","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/generic_utils.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/generic_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e29d3eb7f31a6d05"}},{"code_sha256_prefix":"5b6c17ce2ecdeb80","entry":"make_distinct","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_w2v_ner.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_w2v_ner.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5b6c17ce2ecdeb80"}},{"code_sha256_prefix":"400008f29f6e7364","entry":"parse_result","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_w2v.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_w2v.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"400008f29f6e7364"}},{"code_sha256_prefix":"815aaded7a9eae16","entry":"safe_divide","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_utils_ner.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_utils_ner.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"815aaded7a9eae16"}},{"code_sha256_prefix":"c6afd1a15980c103","entry":"undetected_indices","repo":"asappresearch/slue-toolkit","repo_kind":"official","path":"slue_toolkit/eval/eval_utils_nel.py","file_url":"https://github.com/asappresearch/slue-toolkit/blob/HEAD/slue_toolkit/eval/eval_utils_nel.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c6afd1a15980c103"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}