{"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/sandglasset-a-light-multi-granularity-self","title":"Sandglasset: A Light Multi-Granularity Self-attentive Network For Time-Domain Speech Separation","arxiv_id":"2103.00819","date":"2021-03-01","proceeding":null,"authors":["Max W. Y. Lam","Jun Wang","Dan Su","Dong Yu"],"abstract":"One of the leading single-channel speech separation (SS) models is based on a TasNet with a dual-path segmentation technique, where the size of each segment remains unchanged throughout all layers. In contrast, our key finding is that multi-granularity features are essential for enhancing contextual modeling and computational efficiency. We introduce a self-attentive network with a novel sandglass-shape, namely Sandglasset, which advances the state-of-the-art (SOTA) SS performance at significantly smaller model size and computational cost. Forward along each block inside Sandglasset, the temporal granularity of the features gradually becomes coarser until reaching half of the network blocks, and then successively turns finer towards the raw signal level. We also unfold that residual connections between features with the same granularity are critical for preserving information after passing through the bottleneck layer. Experiments show our Sandglasset with only 2.3M parameters has achieved the best results on two benchmark SS datasets -- WSJ0-2mix and WSJ0-3mix, where the SI-SNRi scores have been improved by absolute 0.8 dB and 2.4 dB, respectively, comparing to the prior SOTA results.","url_abs":"https://arxiv.org/abs/2103.00819v2","url_pdf":"https://arxiv.org/pdf/2103.00819v2.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":"sandglasset-a-light-multi-granularity-self","repo_url":"https://github.com/SoulProficiency/speechseparation-Sandglasset","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"sandglasset-a-light-multi-granularity-self","repo_url":"https://github.com/Zhongyang-debug/Sandglasset-A-Light-Multi-Granularity-Self-Attentive-Network-For-Time-Domain-Speech-Separation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"speech-separation","task_name":"Speech Separation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-separation-on-wsj0-2mix","task":"Speech Separation","dataset":"WSJ0-2mix","model":"Sandglasset","rank_in_archive_order":22,"of":40,"metrics":{"SI-SDRi":"21.0"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-wsj0-3mix","task":"Speech Separation","dataset":"WSJ0-3mix","model":"Sandglasset","rank_in_archive_order":8,"of":9,"metrics":{"SI-SDRi":"17.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2103.00819","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2103.00819"}},"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/Zhongyang-debug/Sandglasset-A-Light-Multi-Granularity-Self-Attentive-Network-For-Time-Domain-Speech-Separation","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/SoulProficiency/speechseparation-Sandglasset","reach":null}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":1,"ran":1,"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":1,"samples":[{"code_sha256_prefix":"a65ff6d7fe19c9c8","entry":"overlap_and_add","repo":"SoulProficiency/speechseparation-Sandglasset","repo_kind":"listed","path":"Sandglasset.py","file_url":"https://github.com/SoulProficiency/speechseparation-Sandglasset/blob/HEAD/Sandglasset.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a65ff6d7fe19c9c8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}