SonicMaster: Towards Controllable All-in-One Music Restoration and Mastering

Music recordings often suffer from audio quality issues such as excessive reverberation, distortion, clipping, tonal imbalances, and a narrowed stereo image, especially when created in non-professional settings without specialized equipment or expertise. These problems are typically corrected using separate specialized tools and manual adjustments. In this paper, we introduce SonicMaster, the first unified generative model for music restoration and mastering that addresses a broad spectrum of audio artifacts with text-based control. SonicMaster is conditioned on natural language instructions to apply targeted enhancements, or can operate in an automatic mode for general restoration. To train this model, we construct the SonicMaster dataset, a large dataset of paired degraded and high-quality tracks by simulating common degradation types with nineteen degradation functions belonging to five enhancements groups: equalization, dynamics, reverb, amplitude, and stereo. Our approach leverages a flow-matching generative training paradigm to learn an audio transformation that maps degraded inputs to their cleaned, mastered versions guided by text prompts. Objective audio quality metrics demonstrate that SonicMaster significantly improves sound quality across all artifact categories. Furthermore, subjective listening tests confirm that listeners prefer SonicMaster's enhanced outputs over other baselines. The model and demo samples are available through https://github.com/AMAAI-Lab/SonicMaster.

Music recordings often suffer from poor audio quality, especially when created by amateur musicians using limited equipment or recording in unsuitable environments such as echoey or noisy rooms. Improving these recordings usually requires expensive tools and expert knowledge in music production and audio engineering. In this work, we introduce SonicMaster, an artificial intelligence system designed to help users improve the quality of their music recordings more easily. The system can enhance music in several different ways, such as reducing unwanted audio problems or improving the overall sound quality. Users can either describe the changes they want using natural language, or allow the system to enhance the music automatically, reducing the need for specialized music-production terminology and technical expertise. To support this research, we also created a new dataset for training and evaluating the system. Our experiments show that SonicMaster is capable of performing a wide range of music enhancement tasks effectively. To encourage further research and accessibility, we publicly release the model, dataset, and an interactive demonstration.