Abstract Sound Fusion with Unconditional Inversion Models
This work addresses sound synthesis for audio applications, but appears incremental as it builds on existing inversion techniques.
The paper tackles the problem of abstract sound fusion by developing novel SDE and ODE inversion models based on DPMSolver++ samplers, which reverse the sampling process to eliminate circular dependencies from noise prediction terms, enabling controllable synthesis without prompt conditioning.
An abstract sound is defined as a sound that does not disclose identifiable real-world sound events to a listener. Sound fusion aims to synthesize an original sound and a reference sound to generate a novel sound that exhibits auditory features beyond mere additive superposition of the sound constituents. To achieve this fusion, we employ inversion techniques that preserve essential features of the original sample while enabling controllable synthesis. We propose novel SDE and ODE inversion models based on DPMSolver++ samplers that reverse the sampling process by configuring model outputs as constants, eliminating circular dependencies incurred by noise prediction terms. Our inversion approach requires no prompt conditioning while maintaining flexible guidance during sampling.