End-to-End Historical Music Restoration in Latent Space

arXiv:2610.00607 · eess.AS, cs.LG, cs.SD, eess.SP · Submitted 2026-09-30 · Read on arXiv

eess.AS, cs.LG, cs.SD, eess.SP

Submitted: 2026-09-30

Updated: 2026-09-30

Code: https://github.com/SonyResearch/historical_

Terminology

• Latent Space — 1× in this paper · explained in 7 episodes

Sources

• A2SB: Audio-to-Audio Schrodinger Bridges
• LatentFlowSR: High-Fidelity Audio Super-Resolution via Noise-Robust Latent Flow Matching
• Aligning Text-to-Music Evaluation with Human Preferences
• SAME: A Semantically-Aligned Music Autoencoder
• Meta Audiobox Aesthetics: Unified Automatic Quality Assessment for Speech, Music, and Sound

Related papers

• Rethinking Entropy Allocation in LLM-based ASR: Understanding the Dynamics between Speech Encoders and LLMs
• X-VC: Zero-shot Streaming Voice Conversion in Codec Space
• Autoregressive Guidance of Deep Spatially Selective Filters using Bayesian Tracking for Efficient Extraction of Moving Speakers
• LiveAnon: Improving Privacy and Utility in Streaming Speaker Anonymization with Codec Language Models
• EarLog: Opportunistic Ear Physiology Sensing toward Personalized Hearing Risk Awareness
• VM-ARRAYDPS: Virtual Microphone Augmented Diffusion Posterior Sampling for Unsupervised Blind Speech Separation