arXiv：cs.LG· Yuhan Liu, Yuxuan Ou, Ruoxi Su, Mohamed Ahmed Zaki, Yunbo Long·· 5 小时前AI 评分36
ParaGeo：将副语言变异分解为共享潜在几何空间
ParaGeo: Decomposing Paralinguistic Variation into a Shared Latent Geometry
AI 导读
ParaGeo 提出一种在冻结语音语言模型中基于匹配内容的副语言变异分解方法，将合成的音频 token 以固定聆听提示词重放，并把池化的 K/V 表示居中投影到共享低维空间。
正文
Abstract:Speech delivery varies with both the requested paralinguistic attribute and the linguistic content. We introduce ParaGeo, a matched-content decomposition of paralinguistic variation in a frozen speech language model. Synthesized audio tokens are replayed with a fixed listening prompt; pooled key/value (K/V) representations are centered and projected into a shared low-dimensional space. Our GLM-4-Voice probe spans 80 requested controls from 12 benchmark families across eight sentences. With a globally fitted calibration basis, content-held-out centroid accuracy using this basis is 9.49% versus a 1.25% permutation baseline; same-label cross-content cosine similarity is 0.285 versus 0.017, and both conditional permutation tests yield p = 0.001. A separate ten-scenario, six-style probe reveals reproducible contrast directions across scenarios. Static, additive, and temporal interventions produce attribute-, layer-, and schedule-dependent response profiles. These results provide a shared coordinate representation for measuring paralinguistic structure and an empirical starting point for latent speech control. Code is available at this https URL.
| Subjects: | Sound (cs.SD); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.03125 [cs.SD] |
| (or arXiv:2610.03125v1 [cs.SD] for this version) | |
| https://doi.org/10.48550/arXiv.2610.03125 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuhan Liu [view email]
[v1]
Fri, 2 Oct 2026 10:46:16 UTC (1,732 KB)
来源：arXiv：cs.LG · arxiv.org