Montesinos García, Juan FelipeKadandale, Venkatesh S.Haro Ortega, Gloria2025-03-272025-03-272022Montesinos JF, Kadandale VS, Haro G. VoViT: low latency graph-based audio-visual voice separation transformer. In: Avidan S, Brostow G, Cissé M, Maria Farinella G, Hassner T, editors. 17th European Conference on Computer Vision Part XVIII (ECCV 2022); 2022 October 23-7; Tel Aviv, Israel. Cham: Springer Verlag; 2022. p.310-26. (LNCS; no. 13678). DOI: 10.1007/978-3-031-19836-6_18http://hdl.handle.net/10230/70025This paper presents an audio-visual approach for voice separation which produces state-of-the-art results at a low latency in two scenarios: speech and singing voice. The model is based on a two-stage network. Motion cues are obtained with a lightweight graph convolutional network that processes face landmarks. Then, both audio and motion features are fed to an audio-visual transformer which produces a fairly good estimation of the isolated target source. In a second stage, the predominant voice is enhanced with an audio-only network. We present different ablation studies and comparison to state-of-the-art methods. Finally, we explore the transferability of models trained for speech separation in the task of singing voice separation. The demos, code, and weights are available in https://ipcv.github.io/VoViT/.application/pdfeng© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 S. Avidan et al. (Eds.): ECCV 2022, LNCS 13697, pp. 310–326, 2022. https://doi.org/10.1007/978-3-031-19836-6_18info:eu-repo/semantics/conferenceObject10.1007/978-3-031-19836-6_18Audio-visualSource separationSpeechSinging voiceinfo:eu-repo/semantics/openAccess