Audio Source Separation

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

2 leaderboard tables shown for this task, 2 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted.

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. A dash means no Syntology record for that paper, not a recorded non-run. Read from the graph 2026-09-24.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

15 datasets whose archive record lists this task, ordered by the archive's paper count.

Subtasks archive 2025-07-28

3 subtasks in the archive's task tree.

Most implemented papers archive 2025-07-28

30 shown of 54 papers with code (112 tagged with this task in all), ordered by repositories listed in the archive, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers without a page here are shown as plain text.

• Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation
  8 Jun 2018 10 repositories listed Syntology ran 1 of 19 samples · 18 unverified · 1 pointer-only (licence)
  Models for audio source separation usually operate on the magnitude spectrum, which ignores phase information and makes separation performance dependant on hyper-parameters for the spectral front-end.
• Multi-scale Multi-band DenseNets for Audio Source Separation
  29 Jun 2017 5 repositories listed
  This paper deals with the problem of audio source separation.
• Sudo rm -rf: Efficient Networks for Universal Audio Source Separation
  14 Jul 2020 4 repositories listed Syntology ran 1 of 10 samples · 9 unverified
  In this paper, we present an efficient neural network for end-to-end general purpose audio source separation.
• The Cocktail Fork Problem: Three-Stem Audio Separation for Real-World Soundtracks
  19 Oct 2021 3 repositories listed Syntology ran 0 of 4 samples · 4 unverified
  The cocktail party problem aims at isolating any source of interest within a complex acoustic scene, and has long inspired audio source separation research.
• Compute and memory efficient universal sound source separation
  3 Mar 2021 3 repositories listed
  Recent progress in audio source separation lead by deep learning has enabled many neural network models to provide robust solutions to this fundamental estimation problem.
• Co-Separating Sounds of Visual Objects
  16 Apr 2019 3 repositories listed
  Learning how objects sound from video is challenging, since they often heavily overlap in a single audio channel.
• Improved Speech Enhancement with the Wave-U-Net
  27 Nov 2018 3 repositories listed
  We study the use of the Wave-U-Net architecture for speech enhancement, a model introduced by Stoller et al for the separation of music vocals and accompaniment.
• Adversarial Semi-Supervised Audio Source Separation applied to Singing Voice Extraction
  31 Oct 2017 3 repositories listed Syntology ran 0 of 13 samples · 13 unverified
  Based on this idea, we drive the separator towards outputs deemed as realistic by discriminator networks that are trained to tell apart real from separator samples.
• Unsupervised Music Source Separation Using Differentiable Parametric Source Models
  24 Jan 2022 2 repositories listed
  Integrating domain knowledge in the form of source models into a data-driven method leads to high data efficiency: the proposed approach achieves good separation quality even when trained on less than three minutes of…
• Directional Sparse Filtering using Weighted Lehmer Mean for Blind Separation of Unbalanced Speech Mixtures
  30 Jan 2021 2 repositories listed
  In blind source separation of speech signals, the inherent imbalance in the source spectrum poses a challenge for methods that rely on single-source dominance for the estimation of the mixing matrix.
• Conditioned-U-Net: Introducing a Control Mechanism in the U-Net for Multiple Source Separations
  2 Jul 2019 2 repositories listed Syntology ran 1 of 1 samples · 0 unverified · 1 pointer-only (licence)
  The input vector is embedded to obtain the parameters that control Feature-wise Linear Modulation (FiLM) layers.
• Learning to Separate Object Sounds by Watching Unlabeled Video
  5 Apr 2018 2 repositories listed
  Our work is the first to learn audio source separation from large-scale "in the wild" videos containing multiple audio sources per video.
• Towards Reliable Objective Evaluation Metrics for Generative Singing Voice Separation Models
  15 Jul 2025 1 repository listed
  Traditional Blind Source Separation Evaluation (BSS-Eval) metrics were originally designed to evaluate linear audio source separation models based on methods such as time-frequency masking.
• Training-Free Multi-Step Audio Source Separation
  26 May 2025 1 repository listed
  Our empirical results demonstrate that our multi-step separation approach consistently outperforms one-step inference across both speech enhancement and music source separation tasks, and can achieve scaling performance…
• Leveraging LLM and Text-Queried Separation for Noise-Robust Sound Event Detection
  2 Nov 2024 1 repository listed
  Sound Event Detection (SED) is challenging in noisy environments where overlapping sounds obscure target events.
• Exploring Text-Queried Sound Event Detection with Audio Source Separation
  20 Sep 2024 1 repository listed Syntology ran 8 of 8 samples · 0 unverified
  To address this issue, we propose the text-queried SED (TQ-SED) framework.
• Unsupervised Composable Representations for Audio
  19 Aug 2024 1 repository listed
  We demonstrate that our framework, used with diffusion models, naturally addresses the task of unsupervised audio source separation, showing that our model is able to perform high-quality separation.
• Facing the Music: Tackling Singing Voice Separation in Cinematic Audio Source Separation
  7 Aug 2024 1 repository listed
  In this work, we demonstrate a very straightforward extension of the dedicated-decoder Bandit and query-based single-decoder Banquet models to a four-stem problem, treating non-musical dialogue, instrumental music,…
• Remastering Divide and Remaster: A Cinematic Audio Source Separation Dataset with Multilingual Support
  9 Jul 2024 1 repository listed
  Cinematic audio source separation (CASS), as a problem of extracting the dialogue, music, and effects stems from their mixture, is a relatively new subtask of audio source separation.
• A Stem-Agnostic Single-Decoder System for Music Source Separation Beyond Four Stems
  26 Jun 2024 1 repository listed Syntology ran 4 of 4 samples · 0 unverified
  Of the very few current systems that support source separation beyond this setup, most continue to rely on an inflexible decoder setup that can only support a fixed pre-defined set of stems.
• Spectral Mapping of Singing Voices: U-Net-Assisted Vocal Segmentation
  30 May 2024 1 repository listed
  The combination of frequency-axis normalization with Min/Max scaling and the Mean Absolute Error (MAE) loss function achieved the highest Source-to-Distortion Ratio (SDR) of 7.
• A Generalized Bandsplit Neural Network for Cinematic Audio Source Separation
  5 Sep 2023 1 repository listed
  Cinematic audio source separation is a relatively new subtask of audio source separation, with the aim of extracting the dialogue, music, and effects stems from their mixture.
• Separate Anything You Describe
  9 Aug 2023 1 repository listed Syntology ran 8 of 9 samples · 1 unverified
  In this work, we introduce AudioSep, a foundation model for open-domain audio source separation with natural language queries.
• Learning Audio-Visual Dynamics Using Scene Graphs for Audio Source Separation
  29 Oct 2022 1 repository listed Syntology ran 1 of 2 samples · 1 unverified
  In this paper, we propose to use this connection between audio and visual dynamics for solving two challenging tasks simultaneously, namely: (i) separating audio sources from a mixture using visual cues, and (ii)…
• Deep Audio Waveform Prior
  21 Jul 2022 1 repository listed
  A network with relevant deep priors is likely to generate a cleaner version of the signal before converging on the corrupted signal.
• Separate What You Describe: Language-Queried Audio Source Separation
  28 Mar 2022 1 repository listed Syntology ran 3 of 3 samples · 0 unverified · 3 pointer-only (licence)
  In this paper, we introduce the task of language-queried audio source separation (LASS), which aims to separate a target source from an audio mixture based on a natural language query of the target source (e.
• Zero-shot Audio Source Separation through Query-based Learning from Weakly-labeled Data
  15 Dec 2021 1 repository listed Syntology ran 3 of 10 samples · 7 unverified
  Our approach uses a single model for source separation of multiple sound types, and relies solely on weakly-labeled data for training.
• Hybrid Neural Networks for On-device Directional Hearing
  11 Dec 2021 1 repository listed
  On-device directional hearing requires audio source separation from a given direction while achieving stringent human-imperceptible latency requirements.
• Transfer Learning with Jukebox for Music Source Separation
  28 Nov 2021 1 repository listed
  In this work, we demonstrate how a publicly available, pre-trained Jukebox model can be adapted for the problem of audio source separation from a single mixed audio channel.
• Unsupervised Source Separation By Steering Pretrained Music Models
  25 Oct 2021 1 repository listed
  We showcase an unsupervised method that repurposes deep models trained for music generation and music tagging for audio source separation, without any retraining.

Syntology lines on 11 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced. Read from the graph 2026-09-24.