{"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/wave-u-net-a-multi-scale-neural-network-for","title":"Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation","arxiv_id":"1806.03185","date":"2018-06-08","proceeding":null,"authors":["Daniel Stoller","Sebastian Ewert","Simon Dixon"],"abstract":"Models for audio source separation usually operate on the magnitude spectrum,\nwhich ignores phase information and makes separation performance dependant on\nhyper-parameters for the spectral front-end. Therefore, we investigate\nend-to-end source separation in the time-domain, which allows modelling phase\ninformation and avoids fixed spectral transformations. Due to high sampling\nrates for audio, employing a long temporal input context on the sample level is\ndifficult, but required for high quality separation results because of\nlong-range temporal correlations. In this context, we propose the Wave-U-Net,\nan adaptation of the U-Net to the one-dimensional time domain, which repeatedly\nresamples feature maps to compute and combine features at different time\nscales. We introduce further architectural improvements, including an output\nlayer that enforces source additivity, an upsampling technique and a\ncontext-aware prediction framework to reduce output artifacts. Experiments for\nsinging voice separation indicate that our architecture yields a performance\ncomparable to a state-of-the-art spectrogram-based U-Net architecture, given\nthe same data. Finally, we reveal a problem with outliers in the currently used\nSDR evaluation metrics and suggest reporting rank-based statistics to alleviate\nthis problem.","url_abs":"http://arxiv.org/abs/1806.03185v1","url_pdf":"http://arxiv.org/pdf/1806.03185v1.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":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/f90/Wave-U-Net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/MattSegal/speech-enhancement","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/ShichengChen/WaveNetSeparateAudio","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/ShichengChen/WaveUNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/craigmacartney/Wave-U-Net-For-Speech-Enhancement","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/f90/Wave-U-Net-Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/haoxiangsnr/Wave-U-Net-for-Speech-Enhancement","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/jperezlapillo/Hyper-Wave-U-Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/lucas-dunker/stem-separator-amt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"wave-u-net-a-multi-scale-neural-network-for","repo_url":"https://github.com/satvik-venkatesh/Wave-U-net-TF2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"audio-source-separation","task_name":"Audio Source Separation"},{"task_slug":"music-source-separation","task_name":"Music Source Separation"}],"methods":[{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"u-net","method_name":"U-Net"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/music-source-separation-on-musdb18","task":"Music Source Separation","dataset":"MUSDB18","model":"STL2","rank_in_archive_order":27,"of":27,"metrics":{"SDR (avg)":"3.23","SDR (bass)":"3.21","SDR (drums)":"4.22","SDR (other)":"2.25","SDR (vocals)":"3.25"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.03185","atlas_url":"https://app.syntology.ai/?focus=1806.03185","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1806.03185"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/craigmacartney/Wave-U-Net-For-Speech-Enhancement","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/haoxiangsnr/Wave-U-Net-for-Speech-Enhancement","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/f90/Wave-U-Net-Pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MattSegal/speech-enhancement","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jperezlapillo/Hyper-Wave-U-Net","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/satvik-venkatesh/Wave-U-net-TF2","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lucas-dunker/stem-separator-amt","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/f90/Wave-U-Net","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ShichengChen/WaveNetSeparateAudio","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ShichengChen/WaveUNet","reach":null}],"summary":{"ran":1,"ran_fixture":1,"unverified":17},"by_repo_kind":{"official":{"samples":9,"ran":0,"repositories":1},"listed":{"samples":10,"ran":2,"repositories":4}},"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":"78f6f864d89c2756","entry":"build_sinc_filter","repo":"f90/Wave-U-Net-Pytorch","repo_kind":"listed","path":"model/resample.py","file_url":"https://github.com/f90/Wave-U-Net-Pytorch/blob/HEAD/model/resample.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"78f6f864d89c2756"}},{"code_sha256_prefix":"19db27f0a2d1b55c","entry":"subsample_chunk_random","repo":"MattSegal/speech-enhancement","repo_kind":"listed","path":"src/datasets/speech/noisy_librispeech/librispeech_dataset.py","file_url":"https://github.com/MattSegal/speech-enhancement/blob/HEAD/src/datasets/speech/noisy_librispeech/librispeech_dataset.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"19db27f0a2d1b55c"}},{"code_sha256_prefix":"2803fec1aab36834","entry":"centre_crop","repo":"f90/Wave-U-Net-Pytorch","repo_kind":"listed","path":"model/crop.py","file_url":"https://github.com/f90/Wave-U-Net-Pytorch/blob/HEAD/model/crop.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":"2803fec1aab36834"}},{"code_sha256_prefix":"950864364ec0b8d2","entry":"compute_loss","repo":"f90/Wave-U-Net-Pytorch","repo_kind":"listed","path":"model/utils.py","file_url":"https://github.com/f90/Wave-U-Net-Pytorch/blob/HEAD/model/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":"950864364ec0b8d2"}},{"code_sha256_prefix":"7ef7210907a0d9be","entry":"compute_mean_metrics","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Evaluate.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Evaluate.py","link_basis":"first_harvest_node","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":"7ef7210907a0d9be"}},{"code_sha256_prefix":"f3054ff75a9faf20","entry":"crop_and_concat","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Utils.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Utils.py","link_basis":"first_harvest_node","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":"f3054ff75a9faf20"}},{"code_sha256_prefix":"404b97b66e233e0d","entry":"getNumParams","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Utils.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Utils.py","link_basis":"first_harvest_node","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":"404b97b66e233e0d"}},{"code_sha256_prefix":"e119d722ad3969f8","entry":"getTrainableVariables","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Utils.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Utils.py","link_basis":"first_harvest_node","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":"e119d722ad3969f8"}},{"code_sha256_prefix":"b293995d32808cdd","entry":"get_lr","repo":"f90/Wave-U-Net-Pytorch","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/f90/Wave-U-Net-Pytorch/blob/HEAD/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":"b293995d32808cdd"}},{"code_sha256_prefix":"d8372ce707c3095c","entry":"get_padding","repo":"satvik-venkatesh/Wave-U-net-TF2","repo_kind":"listed","path":"wave_u_net.py","file_url":"https://github.com/satvik-venkatesh/Wave-U-net-TF2/blob/HEAD/wave_u_net.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":"d8372ce707c3095c"}},{"code_sha256_prefix":"7fd1fa630f0e11bd","entry":"independent_outputs","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Models/OutputLayer.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Models/OutputLayer.py","link_basis":"first_harvest_node","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":"7fd1fa630f0e11bd"}},{"code_sha256_prefix":"aafe9da12cdcd105","entry":"initialize_config","repo":"haoxiangsnr/Wave-U-Net-for-Speech-Enhancement","repo_kind":"listed","path":"util/utils.py","file_url":"https://github.com/haoxiangsnr/Wave-U-Net-for-Speech-Enhancement/blob/HEAD/util/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":"aafe9da12cdcd105"}},{"code_sha256_prefix":"0af32c8aa2b13465","entry":"learned_interpolation_layer","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Models/InterpolationLayer.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Models/InterpolationLayer.py","link_basis":"first_harvest_node","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":"0af32c8aa2b13465"}},{"code_sha256_prefix":"8fdba4f9ae28cbe8","entry":"load_checkpoint","repo":"haoxiangsnr/Wave-U-Net-for-Speech-Enhancement","repo_kind":"listed","path":"util/utils.py","file_url":"https://github.com/haoxiangsnr/Wave-U-Net-for-Speech-Enhancement/blob/HEAD/util/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":"8fdba4f9ae28cbe8"}},{"code_sha256_prefix":"abf55a235181f2e0","entry":"load_model","repo":"f90/Wave-U-Net-Pytorch","repo_kind":"listed","path":"model/utils.py","file_url":"https://github.com/f90/Wave-U-Net-Pytorch/blob/HEAD/model/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":"abf55a235181f2e0"}},{"code_sha256_prefix":"773e07dfbc4df2db","entry":"take_all_snippets","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Datasets.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Datasets.py","link_basis":"first_harvest_node","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":"773e07dfbc4df2db"}},{"code_sha256_prefix":"6950f34674f12037","entry":"take_random_snippets","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Datasets.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Datasets.py","link_basis":"first_harvest_node","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":"6950f34674f12037"}},{"code_sha256_prefix":"b8955c9d5cd1a2ed","entry":"take_snippets_at_pos","repo":"f90/Wave-U-Net","repo_kind":"official","path":"Datasets.py","file_url":"https://github.com/f90/Wave-U-Net/blob/HEAD/Datasets.py","link_basis":"first_harvest_node","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":"b8955c9d5cd1a2ed"}},{"code_sha256_prefix":"8efc19a229bb0bc7","entry":"wave_u_net","repo":"satvik-venkatesh/Wave-U-net-TF2","repo_kind":"listed","path":"wave_u_net.py","file_url":"https://github.com/satvik-venkatesh/Wave-U-net-TF2/blob/HEAD/wave_u_net.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":"8efc19a229bb0bc7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}