{"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/an-embedded-segmental-k-means-model-for","title":"An embedded segmental K-means model for unsupervised segmentation and clustering of speech","arxiv_id":"1703.08135","date":"2017-03-23","proceeding":null,"authors":["Herman Kamper","Karen Livescu","Sharon Goldwater"],"abstract":"Unsupervised segmentation and clustering of unlabelled speech are core\nproblems in zero-resource speech processing. Most approaches lie at\nmethodological extremes: some use probabilistic Bayesian models with\nconvergence guarantees, while others opt for more efficient heuristic\ntechniques. Despite competitive performance in previous work, the full Bayesian\napproach is difficult to scale to large speech corpora. We introduce an\napproximation to a recent Bayesian model that still has a clear objective\nfunction but improves efficiency by using hard clustering and segmentation\nrather than full Bayesian inference. Like its Bayesian counterpart, this\nembedded segmental K-means model (ES-KMeans) represents arbitrary-length word\nsegments as fixed-dimensional acoustic word embeddings. We first compare\nES-KMeans to previous approaches on common English and Xitsonga data sets (5\nand 2.5 hours of speech): ES-KMeans outperforms a leading heuristic method in\nword segmentation, giving similar scores to the Bayesian model while being 5\ntimes faster with fewer hyperparameters. However, its clusters are less pure\nthan those of the other models. We then show that ES-KMeans scales to larger\ncorpora by applying it to the 5 languages of the Zero Resource Speech Challenge\n2017 (up to 45 hours), where it performs competitively compared to the\nchallenge baseline.","url_abs":"http://arxiv.org/abs/1703.08135v2","url_pdf":"http://arxiv.org/pdf/1703.08135v2.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":"an-embedded-segmental-k-means-model-for","repo_url":"https://github.com/kamperh/bucktsong_eskmeans","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"an-embedded-segmental-k-means-model-for","repo_url":"https://github.com/kamperh/eskmeans","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"bayesian-inference","task_name":"Bayesian Inference"},{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.08135","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1703.08135"}},"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/kamperh/bucktsong_eskmeans","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kamperh/eskmeans","reach":null}],"summary":{"ran_fixture":1,"ran_draft_wrong":2,"unverified":12},"by_repo_kind":{"listed":{"samples":15,"ran":3,"repositories":2}},"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":3,"samples":[{"code_sha256_prefix":"49bb6e6331d17d66","entry":"forward_backward_kmeans_viterbi","repo":"kamperh/eskmeans","repo_kind":"listed","path":"eskmeans/eskmeans_wordseg.py","file_url":"https://github.com/kamperh/eskmeans/blob/HEAD/eskmeans/eskmeans_wordseg.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"49bb6e6331d17d66"}},{"code_sha256_prefix":"5d35604b2fba6f41","entry":"local_segment_only_utts","repo":"kamperh/eskmeans","repo_kind":"listed","path":"eskmeans/eskmeans_wordseg.py","file_url":"https://github.com/kamperh/eskmeans/blob/HEAD/eskmeans/eskmeans_wordseg.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"5d35604b2fba6f41"}},{"code_sha256_prefix":"3c99df048b9ea79d","entry":"process_embeddings","repo":"kamperh/eskmeans","repo_kind":"listed","path":"eskmeans/eskmeans_wordseg.py","file_url":"https://github.com/kamperh/eskmeans/blob/HEAD/eskmeans/eskmeans_wordseg.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"3c99df048b9ea79d"}},{"code_sha256_prefix":"9acae200c1954ee9","entry":"average_precision","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"downsample/samediff.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/downsample/samediff.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":"9acae200c1954ee9"}},{"code_sha256_prefix":"744a7b79ce29df7b","entry":"downsample_utterance","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"downsample/downsample_dense.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/downsample/downsample_dense.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":"744a7b79ce29df7b"}},{"code_sha256_prefix":"f42377be880d15f6","entry":"extract_vad","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"features/features.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/features/features.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":"f42377be880d15f6"}},{"code_sha256_prefix":"76167a80296f74b2","entry":"generate_matches_array","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"downsample/samediff.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/downsample/samediff.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":"76167a80296f74b2"}},{"code_sha256_prefix":"bd4d693115d65d6a","entry":"get_durations_dict","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"segmentation/get_data_downsample.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/segmentation/get_data_downsample.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":"bd4d693115d65d6a"}},{"code_sha256_prefix":"e263a5f1a087512f","entry":"get_seglist_from_landmarks","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"downsample/get_seglist.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/downsample/get_seglist.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":"e263a5f1a087512f"}},{"code_sha256_prefix":"0bfe2243a58f7b18","entry":"get_tokens_from_fa","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"segmentation/get_data_downsample.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/segmentation/get_data_downsample.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":"0bfe2243a58f7b18"}},{"code_sha256_prefix":"c03ab624cd19df94","entry":"get_vec_ids_dict","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"segmentation/get_data_downsample.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/segmentation/get_data_downsample.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":"c03ab624cd19df94"}},{"code_sha256_prefix":"e09ac618a9f85288","entry":"mean_average_precision","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"downsample/samediff.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/downsample/samediff.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":"e09ac618a9f85288"}},{"code_sha256_prefix":"fa13b362a551768b","entry":"read_vad_from_fa","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"features/utils.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/features/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":"fa13b362a551768b"}},{"code_sha256_prefix":"38722d202e76b2cb","entry":"strip_nonvad","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"features/utils.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/features/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":"38722d202e76b2cb"}},{"code_sha256_prefix":"eaa884c8ef9ed624","entry":"uttlabel_to_uttkey","repo":"kamperh/bucktsong_eskmeans","repo_kind":"listed","path":"features/utils.py","file_url":"https://github.com/kamperh/bucktsong_eskmeans/blob/HEAD/features/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":"eaa884c8ef9ed624"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}