MoVE: Translating Laughter and Tears via Mixture of Vocalization Experts in Speech-to-Speech Translation
Recent Speech-to-Speech Translation (S2ST) systems achieve strong semantic accuracy yet consistently strip away non-verbal vocalizations (NVs), such as laughter and crying that convey pragmatic intent, which severely limits real-world utility. We address this via three contributions. First, we propose a synthesis pipeline for building scalable expressive datasets to overcome the data scarcity limitation. Second, we propose MoVE, a Mixture-of-LoRA-Experts architecture with expressive-specialized adapters and a soft-weighting router that blends experts for capturing hybrid expressive states. Third, we show pretrained AudioLLMs enable striking data efficiency: 30 minutes of curated data is e...
| Authors | Szu-Chi Chen, I-Ning Tsai, Yi-Cheng Lin et al. |
|---|---|
| Primary category | cs.CL |
| Published | 19 April 2026 |
| Last updated at source | 19 April 2026 |
Connected in the knowledge graph
- 47z/MoVEDataset · cited by
Models and datasets referencing this paper
- 47z/MoVEDataset
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.