SOTAVerified

Audio Source Separation

Audio Source Separation is the process of separating a mixture (e.g. a pop band recording) into isolated sounds from individual sources (e.g. just the lead vocals).

Source: Model selection for deep audio source separation via clustering analysis

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 112 papers

TitleStatusHype
Separate Anything You DescribeCode3
Training-Free Multi-Step Audio Source SeparationCode2
A Stem-Agnostic Single-Decoder System for Music Source Separation Beyond Four StemsCode2
Unsupervised Source Separation via Bayesian Inference in the Latent DomainCode1
Unsupervised Source Separation By Steering Pretrained Music ModelsCode1
Unsupervised Composable Representations for AudioCode1
The Cone of Silence: Speech Separation by LocalizationCode1
The Cocktail Fork Problem: Three-Stem Audio Separation for Real-World SoundtracksCode1
Time-Domain Audio Source Separation Based on Wave-U-Net Combined with Discrete Wavelet TransformCode1
Hybrid Neural Networks for On-device Directional HearingCode1
Leveraging LLM and Text-Queried Separation for Noise-Robust Sound Event DetectionCode1
AutoClip: Adaptive Gradient Clipping for Source Separation NetworksCode1
Unified Gradient Reweighting for Model Biasing with Applications to Source SeparationCode1
Unsupervised Music Source Separation Using Differentiable Parametric Source ModelsCode1
Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source SeparationCode1
OtoWorld: Towards Learning to Separate by Learning to MoveCode1
Remastering Divide and Remaster: A Cinematic Audio Source Separation Dataset with Multilingual SupportCode1
Transfer Learning with Jukebox for Music Source SeparationCode1
Zero-shot Audio Source Separation through Query-based Learning from Weakly-labeled DataCode1
Separate What You Describe: Language-Queried Audio Source SeparationCode1
Spectral Mapping of Singing Voices: U-Net-Assisted Vocal SegmentationCode1
Deep Audio Waveform PriorCode1
Sudo rm -rf: Efficient Networks for Universal Audio Source SeparationCode1
Parallel and Flexible Sampling from Autoregressive Models via Langevin DynamicsCode1
Differentiable Model Compression via Pseudo Quantization NoiseCode1
Directional Sparse Filtering using Weighted Lehmer Mean for Blind Separation of Unbalanced Speech MixturesCode1
Exploring Text-Queried Sound Event Detection with Audio Source SeparationCode1
Facing the Music: Tackling Singing Voice Separation in Cinematic Audio Source SeparationCode1
Compute and memory efficient universal sound source separationCode1
Unsupervised Audio Source Separation using Generative PriorsCode1
A Generalized Bandsplit Neural Network for Cinematic Audio Source SeparationCode1
Multi-Task Audio Source SeparationCode1
Learning to Separate Object Sounds by Watching Unlabeled VideoCode0
Learning Audio-Visual Dynamics Using Scene Graphs for Audio Source SeparationCode0
Training Generative Adversarial Networks from Incomplete Observations using Factorised DiscriminatorsCode0
J-Net: Randomly weighted U-Net for audio source separationCode0
Densely connected multidilated convolutional networks for dense prediction tasksCode0
Sams-Net: A Sliced Attention-based Neural Network for Music Source SeparationCode0
Densely Connected Multi-Dilated Convolutional Networks for Dense Prediction TasksCode0
Sparse Gaussian Process Audio Source Separation Using Spectrum Priors in the Time-DomainCode0
Improved Speech Enhancement with the Wave-U-NetCode0
A Provably Correct and Robust Algorithm for Convolutive Nonnegative Matrix FactorizationCode0
Co-Separating Sounds of Visual ObjectsCode0
Retrieving Signals in the Frequency Domain with Deep Complex ExtractorsCode0
Adversarial Semi-Supervised Audio Source Separation applied to Singing Voice ExtractionCode0
Audio-Visual Scene Analysis with Self-Supervised Multisensory FeaturesCode0
Sampling-Frequency-Independent Audio Source Separation Using Convolution Layer Based on Impulse Invariant MethodCode0
Conditioned-U-Net: Introducing a Control Mechanism in the U-Net for Multiple Source SeparationsCode0
Generalization Challenges for Neural Architectures in Audio Source SeparationCode0
Music source separation conditioned on 3D point cloudsCode0
Show:102550
← PrevPage 1 of 3Next →
All datasetsAudioSetMUSIC (multi-source)

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ST-SED-SEPSDR10.55—Unverified
2Co-SeparationSDR4.26—Unverified
#ModelMetricClaimedVerifiedStatus
1Co-SeparationSAR11.3—Unverified