Latent Diffusion Bridges For Unsupervised Musical Audio Timbre Transfer

Abstract

Music timbre transfer is a challenging task that involves modifying the timbral characteristics of an audio signal while preserving its melodic structure. In this paper, we propose a novel method based on dual diffusion bridges, trained using the CocoChorales Dataset, which consists of unpaired monophonic single-instrument audio data. Each diffusion model is trained on a specific instrument with a Gaussian prior. During inference, a model is designated as the source model to map the input audio to its corresponding Gaussian prior, and another model is designated as the target model to reconstruct the target audio from this Gaussian prior, thereby facilitating timbre transfer. We compare our approach against existing unsupervised timbre transfer models such as VAEGAN and Gaussian Flow Bridges (GFB). Experimental results demonstrate that our method achieves both better Fr\'echet Audio Distance (FAD) and melody preservation, as reflected by lower pitch distances (DPD) compared to VAEGAN a

Stats

• citations2
• S2 citations—
• github stars0
• HF likes0
• heat score3.58
• arxiv keymancusi2024latent

Related papers

• Timbre Transfer With Variational Auto Encoding And Cycle-consistent Adversarial Networks (2021)0.00
• Play As You Like: Timbre-enhanced Multi-modal Music Style Transfer (2018)9.92
• Token-based Audio Inpainting Via Discrete Diffusion (2025)0.00
• Stylus: Repurposing Stable Diffusion For Training-free Music Style Transfer On Mel-spectrograms (2024)0.00
• Schrodinger Bridges Beat Diffusion Models On Text-to-speech Synthesis (2023)0.00
• Conditional Diffusion As Latent Constraints For Controllable Symbolic Music Generation (2025)0.00
• Edmsound: Spectrogram Based Diffusion Models For Efficient And High-quality Audio Synthesis (2023)0.00
• Diffgap: A Lightweight Diffusion Module In Contrastive Space For Bridging Cross-model Gap (2025)3.58