Music Source Restoration with Ensemble Separation and Targeted Reconstruction
This addresses the problem of reversing complex audio production processes for music restoration, but it is incremental as it builds on existing separation and restoration models.
The paper tackles the Music Source Restoration (MSR) Challenge by recovering original stems from mixed and mastered music, achieving second place on the official benchmark with improved metrics over baselines.
The Inaugural Music Source Restoration (MSR) Challenge targets the recovery of original, unprocessed stems from fully mixed and mastered music. Unlike conventional music source separation, MSR requires reversing complex production processes such as equalization, compression, reverberation, and other real-world degradations. To address MSR, we propose a two-stage system. First, an ensemble of pre-trained separation models produces preliminary source estimates. Then a set of pre-trained BSRNN-based restoration models performs targeted reconstruction to refine these estimates. On the official MSR benchmark, our system surpasses the baselines on all metrics, ranking second among all submissions. The code is available at https://github.com/xinghour/Music-source-restoration-CUPAudioGroup