Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.14912 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918187816189952 |
|---|---|
| author | Wang, Kangdi Wu, Zhiyue Zhou, Dinghao Lin, Rui Dai, Junyu Jiang, Tao |
| author_facet | Wang, Kangdi Wu, Zhiyue Zhou, Dinghao Lin, Rui Dai, Junyu Jiang, Tao |
| contents | Variational Autoencoders (VAEs) are essential for large-scale audio tasks like diffusion-based generation. However, existing open-source models often neglect auditory perceptual aspects during training, leading to weaknesses in phase accuracy and stereophonic spatial representation. To address these challenges, we propose εar-VAE, an open-source music signal reconstruction model that rethinks and optimizes the VAE training paradigm. Our contributions are threefold: (i) A K-weighting perceptual filter applied prior to loss calculation to align the objective with auditory perception. (ii) Two novel phase losses: a Correlation Loss for stereo coherence, and a Phase Loss using its derivatives--Instantaneous Frequency and Group Delay--for precision. (iii) A new spectral supervision paradigm where magnitude is supervised by all four Mid/Side/Left/Right components, while phase is supervised only by the LR components. Experiments show εar-VAE at 44.1kHz substantially outperforms leading open-source models across diverse metrics, showing particular strength in reconstructing high-frequency harmonics and the spatial characteristics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14912 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Back to Ear: Perceptually Driven High Fidelity Music Reconstruction Wang, Kangdi Wu, Zhiyue Zhou, Dinghao Lin, Rui Dai, Junyu Jiang, Tao Sound Artificial Intelligence Variational Autoencoders (VAEs) are essential for large-scale audio tasks like diffusion-based generation. However, existing open-source models often neglect auditory perceptual aspects during training, leading to weaknesses in phase accuracy and stereophonic spatial representation. To address these challenges, we propose εar-VAE, an open-source music signal reconstruction model that rethinks and optimizes the VAE training paradigm. Our contributions are threefold: (i) A K-weighting perceptual filter applied prior to loss calculation to align the objective with auditory perception. (ii) Two novel phase losses: a Correlation Loss for stereo coherence, and a Phase Loss using its derivatives--Instantaneous Frequency and Group Delay--for precision. (iii) A new spectral supervision paradigm where magnitude is supervised by all four Mid/Side/Left/Right components, while phase is supervised only by the LR components. Experiments show εar-VAE at 44.1kHz substantially outperforms leading open-source models across diverse metrics, showing particular strength in reconstructing high-frequency harmonics and the spatial characteristics. |
| title | Back to Ear: Perceptually Driven High Fidelity Music Reconstruction |
| topic | Sound Artificial Intelligence |
| url | https://arxiv.org/abs/2509.14912 |