# High-Quality Visually-Guided Sound Separation from Diverse Categories

Source: [High-Quality Visually-Guided Sound Separation from Diverse Categories](<https://www.sota2.com/research/paper/68fa75351e209131cea66710>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:41:00.437Z

Authors: Chao Huang, Susan Liang, Yapeng Tian, Anurag Kumar, Chenliang Xu

Published: 2023-07-31T19:41:49.000Z

## Abstract

We propose DAVIS, a Diffusion-based Audio-VIsual Separation framework that solves the audio-visual sound source separation task through generative learning. Existing methods typically frame sound separation as a mask-based regression problem, achieving significant progress. However, they face limitations in capturing the complex data distribution required for high-quality separation of sounds from diverse categories. In contrast, DAVIS leverages a generative diffusion model and a Separation U-Net to synthesize separated sounds directly from Gaussian noise, conditioned on both the audio mixture and the visual information. With its generative objective, DAVIS is better suited to achieving the goal of high-quality sound separation across diverse sound categories. We compare DAVIS to existing state-of-the-art discriminative audio-visual separation methods on the AVE and MUSIC datasets, and results show that DAVIS outperforms other methods in separation quality, demonstrating the advantages of our framework for tackling the audio-visual source separation task.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Audio Source Separation on MUSIC (test)](<https://www.sota2.com/research/sota/audio-source-separation-on-music-test.md>): Audio Source Separation; MUSIC (test); SDR: 11.61.

- [Audio Source Separation on MoXaRt XR Scenarios (human speech sessions)](<https://www.sota2.com/research/sota/audio-source-separation-on-moxart-xr-scenarios-human-speech-sessions.md>): Audio Source Separation; MoXaRt XR Scenarios (human speech sessions); WER: 0.7259.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2308.00122>)

[PDF](<https://arxiv.org/pdf/2308.00122.pdf>)
