paper-with-me

홈 › Papers

Taming Audio VAEs via Target-KL Regularization

2026-05-16 · Prem Seetharaman, Rithesh Kumar arxiv

Latent diffusion models have emerged as the dominant paradigm for many generation tasks including audio generation such as text-to-audio, text-to-music and text-to-speech. A key component of latent diffusion is an autoencoder (VAE) that compresses high-dimensional signals into a low frame rate continuous representation that is conducive for downstream prediction. Regularizing these VAEs is challenging, as there is a trade-off between over-regularized (poor output quality) and under-regularized (difficult to predict) latent representations. We propose a framework for studying this trade-off through compression and train Audio VAEs at specific bitrates via target-KL regularization. This allows direct comparison to well-studied discrete neural audio codec models, and the construction of rate-distortion curves for audio VAEs. We evaluate the impact of target-KL regularization on text-to-sound generation and find that sweeping compression rates is helpful in identifying the optimal generation setting.

📄 PDF Abstract BibTeX arXiv:2605.17085

Code (0)

등록된 구현이 없습니다.

Tasks

Audio Generation

Similar Papers 제목 키워드 기반

Learning Hierarchical Priors in VAEs

2019-05-13 · NeurIPS 2019 12 · Alexej Klushyn, Nutan Chen, Richard Kurle, Botond Cseke 외

We propose to learn a hierarchical prior in the context of variational autoencoders to avoid the over-regularisation resulting from a standard normal prior distribution. To incentivise an informative latent representatio…

Taming under isoperimetry

2023-11-15 · Iosif Lytras, Sotirios Sabanis

In this article we propose a novel taming Langevin-based scheme called $\mathbf{sTULA}$ to sample from distributions with superlinearly growing log-gradient which also satisfy a Log-Sobolev inequality. We derive non-asym…

SepVAE: a contrastive VAE to separate pathological patterns from healthy ones

2023-07-12 · Robin Louiset, Edouard Duchesnay, Antoine Grigis, Benoit Dufumier 외

Contrastive Analysis VAE (CA-VAEs) is a family of Variational auto-encoders (VAEs) that aims at separating the common factors of variation between a background dataset (BG) (i.e., healthy subjects) and a target dataset (…

F3-Tokenizer: Taming Audio Autoencoder Latents for Understanding and Generation

2026-06-04 · Dinghao Zhou, Xingchen Song, Di Wu, Pengyu Cheng 외 arxiv

Continuous audio autoencoders reconstruct waveforms well but often produce latents with weak structure for understanding, while self-supervised audio encoders capture semantics but are not directly decodable. This mismat…

Prior Learning in Introspective VAEs

2024-08-25 · Ioannis Athanasiadis, Fredrik Lindsten, Michael Felsberg

Variational Autoencoders (VAEs) are a popular framework for unsupervised learning and data generation. A plethora of methods have been proposed focusing on improving VAEs, with the incorporation of adversarial objectives…

Density EstimationImage GenerationRepresentation Learning