{"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/tdfnet-an-efficient-audio-visual-speech","title":"TDFNet: An Efficient Audio-Visual Speech Separation Model with Top-down Fusion","arxiv_id":"2401.14185","date":"2024-01-25","proceeding":null,"authors":["Samuel Pegg","Kai Li","Xiaolin Hu"],"abstract":"Audio-visual speech separation has gained significant traction in recent years due to its potential applications in various fields such as speech recognition, diarization, scene analysis and assistive technologies. Designing a lightweight audio-visual speech separation network is important for low-latency applications, but existing methods often require higher computational costs and more parameters to achieve better separation performance. In this paper, we present an audio-visual speech separation model called Top-Down-Fusion Net (TDFNet), a state-of-the-art (SOTA) model for audio-visual speech separation, which builds upon the architecture of TDANet, an audio-only speech separation method. TDANet serves as the architectural foundation for the auditory and visual networks within TDFNet, offering an efficient model with fewer parameters. On the LRS2-2Mix dataset, TDFNet achieves a performance increase of up to 10\\% across all performance metrics compared with the previous SOTA method CTCNet. Remarkably, these results are achieved using fewer parameters and only 28\\% of the multiply-accumulate operations (MACs) of CTCNet. In essence, our method presents a highly effective and efficient solution to the challenges of speech separation within the audio-visual domain, making significant strides in harnessing visual information optimally.","url_abs":"https://arxiv.org/abs/2401.14185v1","url_pdf":"https://arxiv.org/pdf/2401.14185v1.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":"tdfnet-an-efficient-audio-visual-speech","repo_url":"https://github.com/spkgyk/TDFNet","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-separation","task_name":"Speech Separation"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-separation-on-lrs2","task":"Speech Separation","dataset":"LRS2","model":"TDFNet-large","rank_in_archive_order":2,"of":8,"metrics":{"PESQ":"3.21","SDRi":"15.9","SI-SNRi":"15.8","STOI":"0.949"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs2","task":"Speech Separation","dataset":"LRS2","model":"TDFNet (MHSA + Shared)","rank_in_archive_order":3,"of":8,"metrics":{"PESQ":"3.16","SDRi":"15.2","SI-SNRi":"15.0","STOI":"0.938"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-lrs2","task":"Speech Separation","dataset":"LRS2","model":"TDFNet-small","rank_in_archive_order":8,"of":8,"metrics":{"PESQ":"3.10","SDRi":"13.7","SI-SNRi":"13.6","STOI":"0.931"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}