{"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/speech-denoising-with-deep-feature-losses","title":"Speech Denoising with Deep Feature Losses","arxiv_id":"1806.10522","date":"2018-06-27","proceeding":null,"authors":["Francois G. Germain","Qifeng Chen","Vladlen Koltun"],"abstract":"We present an end-to-end deep learning approach to denoising speech signals\nby processing the raw waveform directly. Given input audio containing speech\ncorrupted by an additive background signal, the system aims to produce a\nprocessed signal that contains only the speech content. Recent approaches have\nshown promising results using various deep network architectures. In this\npaper, we propose to train a fully-convolutional context aggregation network\nusing a deep feature loss. That loss is based on comparing the internal feature\nactivations in a different network, trained for acoustic environment detection\nand domestic audio tagging. Our approach outperforms the state-of-the-art in\nobjective speech quality metrics and in large-scale perceptual experiments with\nhuman listeners. It also outperforms an identical network trained using\ntraditional regression losses. The advantage of the new approach is\nparticularly pronounced for the hardest data with the most intrusive background\nnoise, for which denoising is most needed and most challenging.","url_abs":"http://arxiv.org/abs/1806.10522v1","url_pdf":"http://arxiv.org/pdf/1806.10522v1.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":"speech-denoising-with-deep-feature-losses","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":"speech-denoising-with-deep-feature-losses","repo_url":"https://github.com/alexander-prutko/mil-audio-denoising","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"speech-denoising-with-deep-feature-losses","repo_url":"https://github.com/anicolson/DeepXi","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MPL-2.0"}},{"paper_slug":"speech-denoising-with-deep-feature-losses","repo_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"speech-denoising-with-deep-feature-losses","repo_url":"https://github.com/kuntojirohan/speechdenoising","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"audio-tagging","task_name":"Audio Tagging"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"speech-denoising","task_name":"Speech Denoising"}],"methods":[{"method_slug":"3d-convolution","method_name":"3D Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.10522","atlas_url":"https://app.syntology.ai/?focus=1806.10522","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1806.10522"}},"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/alexander-prutko/mil-audio-denoising","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/anicolson/DeepXi","reach":{"status":"ok","spdx":"MPL-2.0"}},{"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/kuntojirohan/speechdenoising","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":6},"by_repo_kind":{"listed":{"samples":6,"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":"dc46d9580f8807c9","entry":"dilated_to_signal","repo":"francoisgermain/SpeechDenoisingWithDeepFeatureLosses","repo_kind":"listed","path":"helper.py","file_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses/blob/HEAD/helper.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":"dc46d9580f8807c9"}},{"code_sha256_prefix":"892d80718b66b7a2","entry":"featureloss","repo":"francoisgermain/SpeechDenoisingWithDeepFeatureLosses","repo_kind":"listed","path":"model.py","file_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses/blob/HEAD/model.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":"892d80718b66b7a2"}},{"code_sha256_prefix":"2d85eccb9fa56d87","entry":"lossnet","repo":"francoisgermain/SpeechDenoisingWithDeepFeatureLosses","repo_kind":"listed","path":"model.py","file_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses/blob/HEAD/model.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":"2d85eccb9fa56d87"}},{"code_sha256_prefix":"1a2446e08ebe5a2b","entry":"lrelu","repo":"francoisgermain/SpeechDenoisingWithDeepFeatureLosses","repo_kind":"listed","path":"helper.py","file_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses/blob/HEAD/helper.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":"1a2446e08ebe5a2b"}},{"code_sha256_prefix":"cb7de482eccd62b7","entry":"senet","repo":"francoisgermain/SpeechDenoisingWithDeepFeatureLosses","repo_kind":"listed","path":"model.py","file_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses/blob/HEAD/model.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":"cb7de482eccd62b7"}},{"code_sha256_prefix":"5c811d1e2d397c02","entry":"signal_to_dilated","repo":"francoisgermain/SpeechDenoisingWithDeepFeatureLosses","repo_kind":"listed","path":"helper.py","file_url":"https://github.com/francoisgermain/SpeechDenoisingWithDeepFeatureLosses/blob/HEAD/helper.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":"5c811d1e2d397c02"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}