SCAPES：面向环境声音的语义条件自回归先验模型
SCAPES: Semantically Conditioned Autoregressive Prior for Environmental Sounds
浏览论文内容
中文总结 AI 辅助
本文提出轻量高效的生成式音频模型SCAPES，可通过语义控制合成高保真环境声音，仅需单个消费级GPU即可训练，支持语义插值，相关资源已公开。
中文摘要 AI 辅助
随着生成式音频模型复杂度的提升，合成日常声音的计算成本与生态代价已愈发高昂，往往需要工业级规模的资源与海量数据集。本文提出SCAPES：面向环境声音的语义条件自回归先验模型，这是一款轻量、资源高效的生成式模型，旨在通过高级语义控制合成高保真的环境纹理。该模型在神经音频编解码器的连续潜在流形上运行，可规避离散分词固有的刚性结构约束。我们提出一种分段策略，将音频分解为重叠片段，使连续归一化流（CNF）能利用流匹配（Flow Matching）对潜在轨迹的演化进行建模。实验表明，参数量为3600万的SCAPES实例可在有限的未整理数据集上，通过单个消费级GPU完成训练；值得注意的是，训练时长约为源音频时长的两倍即可实现收敛，生成的输出具备高保真度、稳健的长期稳定性与语义一致性。此外，我们展示了该模型实现平滑语义插值的能力，为开放研究与创意声音设计提供了灵活且易用的工具。代码、预训练权重、音频示例及交互式演示已在项目页面公开，链接为this https URL。
英文摘要
As generative audio models grow in complexity, the computational and ecological costs of synthesizing everyday sounds have become increasingly prohibitive, often requiring industrial-scale resources and massive datasets. In this paper, we present SCAPES: a Semantically Conditioned Autoregressive Prior for Environmental Sounds. SCAPES is a lightweight, resource-efficient generative model designed to synthesize high-fidelity environmental textures through high-level semantic control. By operating on the continuous latent manifold of a neural audio codec, our approach bypasses the rigid structural constraints inherent to discrete tokenization. We propose a segmentation strategy that decomposes audio into overlapping segments, enabling a Continuous Normalizing Flow (CNF) to model the evolution of latent trajectories using Flow Matching. Our experiments demonstrate that a 36-million parameter instance of SCAPES can be trained on limited, uncurated datasets using a single consumer-grade GPU. Notably, convergence is achieved after training for approximately twice the source audio duration, yielding high-fidelity outputs with robust long-term stability and semantic consistency. Furthermore, we showcase the model's capacity for smooth semantic interpolation, providing a flexible and accessible tool for open research and creative sound design. Code, pretrained weights, audio examples, and an interactive demo are publicly available on our project page https://cordutie.github.io/projects/scapes.html
