可解码性不等于可控性：连续隐空间 TTS 中的局部与各向异性说话人控制

View PDF HTML (experimental)

Abstract:High linear decodability does not entail causal controllability in continuous-latent text-to-speech (TTS). In VoxCPM2, position-matched probes and equal-norm interventions across 30 speakers and 60 directed dyads expose this separation: reference-audio states at the final residual acoustic language model layer (RALM L7) retain 0.970 probe accuracy, yet aligned edits yield 0/60 target-dominant dyads. Edits at early reference-audio positions produce the opposite pattern. At L0, pair-aligned updates increase target similarity by 0.326 and achieve target dominance in all 60 dyads, against at most 1/60 under any equal-norm direction or position control. Independent speaker encoders, retrieval against held-out speaker prototypes, and double-blind listening confirm that the shifted voice is heard as the intended target. Cross-architecture interventions in this http URL distinguish backbone leverage from direct acoustic control: backbone edits displace the source without target attraction, while x-vector ablation reduces source similarity from 0.849 to 0.006. These findings identify two requirements for training-free speaker control: geometric specificity and a computational route that carries the intended speaker condition into acoustic synthesis.

Subjects:	Sound (cs.SD); Computation and Language (cs.CL); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
Cite as:	arXiv:2610.29345 [cs.SD]
	(or arXiv:2610.29345v1 [cs.SD] for this version)
	https://doi.org/10.48550/arXiv.2610.29345 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yuanzhong Zheng [view email]
[v1] Wed, 21 Oct 2026 07:38:25 UTC (310 KB)