{"ID":5551659,"CreatedAt":"2026-07-02T01:54:51.863792489Z","UpdatedAt":"2026-07-04T13:37:00.247962456Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.00946","arxiv_id":"2607.00946","title":"A Geometric Perspective on Composable Emotion Steering in Text-to-Speech Models","abstract":"While prior work has explored emotion control in hybrid text-to-speech systems, the geometric properties of these modules, and their implications for steerability, remain poorly understood. We present the first comparative study of speech language model (SLM) and conditional flow-matching (CFM) modules as activation steering sites for mixed emotion speech synthesis. We first characterize emotion representations using linear probing and local intrinsic dimensionality (LID), and then evaluate single-site and joint steering for mixed-emotion synthesis. Our results show that SLM offers a clean, low-dimensional emotion-specific subspace with strong speaker--emotion disentanglement, while CFM exhibitspoor cross-speaker generalization due to speaker--emotion entanglement. Joint steering increases emotion intensity but degrades proportional control and speech quality on in-distribution data. These findings provide practical guidance for multi-site activation steering in hybrid TTS systems and highlight the importance of representation geometry in controllable speech generation.","short_abstract":"While prior work has explored emotion control in hybrid text-to-speech systems, the geometric properties of these modules, and their implications for steerability, remain poorly understood. We present the first comparative study of speech language model (SLM) and conditional flow-matching (CFM) modules as activation st...","url_abs":"https://arxiv.org/abs/2607.00946","url_pdf":"https://arxiv.org/pdf/2607.00946v1","authors":"[\"Siyi Wang\",\"James Bailey\",\"Ting Dang\"]","published":"2026-07-01T13:46:16Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.LG\"]","methods":"[\"Language Model\"]","has_code":false}