{"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/rtfs-net-recurrent-time-frequency-modelling","title":"RTFS-Net: Recurrent Time-Frequency Modelling for Efficient Audio-Visual Speech Separation","arxiv_id":"2309.17189","date":"2023-09-29","proceeding":null,"authors":["Samuel Pegg","Kai Li","Xiaolin Hu"],"abstract":"Audio-visual speech separation methods aim to integrate different modalities to generate high-quality separated speech, thereby enhancing the performance of downstream tasks such as speech recognition. Most existing state-of-the-art (SOTA) models operate in the time domain. However, their overly simplistic approach to modeling acoustic features often necessitates larger and more computationally intensive models in order to achieve SOTA performance. In this paper, we present a novel time-frequency domain audio-visual speech separation method: Recurrent Time-Frequency Separation Network (RTFS-Net), which applies its algorithms on the complex time-frequency bins yielded by the Short-Time Fourier Transform. We model and capture the time and frequency dimensions of the audio independently using a multi-layered RNN along each dimension. Furthermore, we introduce a unique attention-based fusion technique for the efficient integration of audio and visual information, and a new mask separation approach that takes advantage of the intrinsic spectral nature of the acoustic features for a clearer separation. RTFS-Net outperforms the prior SOTA method in both inference speed and separation quality while reducing the number of parameters by 90% and MACs by 83%. This is the first time-frequency domain audio-visual speech separation method to outperform all contemporary time-domain counterparts.","url_abs":"https://arxiv.org/abs/2309.17189v4","url_pdf":"https://arxiv.org/pdf/2309.17189v4.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":"rtfs-net-recurrent-time-frequency-modelling","repo_url":"https://github.com/spkgyk/RTFS-Net","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"audio-visual-speech-recognition","task_name":"Audio-Visual Speech Recognition"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-separation","task_name":"Speech Separation"},{"task_slug":"target-speaker-extraction","task_name":"Target Speaker Extraction"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-separation-on-lrs2","task":"Speech Separation","dataset":"LRS2","model":"RTFS-Net-12","rank_in_archive_order":4,"of":8,"metrics":{"SDRi":"15.1","SI-SNRi":"14.9"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs2","task":"Speech Separation","dataset":"LRS2","model":"RTFS-Net-6","rank_in_archive_order":5,"of":8,"metrics":{"SDRi":"14.8","SI-SNRi":"14.6"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs2","task":"Speech Separation","dataset":"LRS2","model":"RTFS-Net-4","rank_in_archive_order":7,"of":8,"metrics":{"SDRi":"14.3","SI-SNRi":"14.1"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs3","task":"Speech Separation","dataset":"LRS3","model":"RTFS-Net-12","rank_in_archive_order":2,"of":5,"metrics":{"SDRi":"17.6","SI-SNRi":"17.5"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs3","task":"Speech Separation","dataset":"LRS3","model":"RTFS-Net-6","rank_in_archive_order":4,"of":5,"metrics":{"SDRi":"17.1","SI-SNRi":"16.9"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs3","task":"Speech Separation","dataset":"LRS3","model":"RTFS-Net-4","rank_in_archive_order":5,"of":5,"metrics":{"SDRi":"15.6","SI-SNRi":"15.5"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-voxceleb2","task":"Speech Separation","dataset":"VoxCeleb2","model":"RTFS-Net-12","rank_in_archive_order":2,"of":5,"metrics":{"SDRi":"13.6","SI-SNRi":"12.4"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-voxceleb2","task":"Speech Separation","dataset":"VoxCeleb2","model":"RTFS-Net-6","rank_in_archive_order":4,"of":5,"metrics":{"SDRi":"12.8","SI-SNRi":"11.8"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-voxceleb2","task":"Speech Separation","dataset":"VoxCeleb2","model":"RTFS-Net-4","rank_in_archive_order":5,"of":5,"metrics":{"SDRi":"12.4","SI-SNRi":"11.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2309.17189","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.17189"}},"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/spkgyk/RTFS-Net","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"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":"7b8df33b830a890a","entry":"MultiModalFusion","repo":"spkgyk/RTFS-Net","repo_kind":"official","path":"src/models/TDAVNet/fusion.py","file_url":"https://github.com/spkgyk/RTFS-Net/blob/HEAD/src/models/TDAVNet/fusion.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":"7b8df33b830a890a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}