Metraipolis › archive › dc_e6ae87a8707a
SEmoEdit: Probing and Harnessing the Editability of Pre-trained Speech Flows — arXiv:2609.34648
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34648
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:02:47.394Z
- SHA-256 of the archived text
f9c0db2b1c2f5022835eca2f9654d26e0a68afb229afdd634713bd4496717644- Stable URL of this page
https://metraipolis.com/documento/dc_e6ae87a8707a
How to cite
SEmoEdit: Probing and Harnessing the Editability of Pre-trained Speech Flows — arXiv:2609.34648 — arXiv (en), https://arxiv.org/abs/2609.34648 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e6ae87a8707a (Metraipolis archive dc_e6ae87a8707a, sha256 f9c0db2b1c2f5022835eca2f9654d26e0a68afb229afdd634713bd4496717644)
BibTeX
@misc{dce6ae87a8707a,
title = {SEmoEdit: Probing and Harnessing the Editability of Pre-trained Speech Flows — arXiv:2609.34648},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34648},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e6ae87a8707a; sha256 f9c0db2b1c2f5022835eca2f9654d26e0a68afb229afdd634713bd4496717644},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34648. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Existing training-based speech emotion editing methods often require substantial task-specific training and can be unstable. This motivates us to investigate whether the pretrained generative dynamics of large-scale text-to-speech (TTS) models can be directly manipulated for training-free emotion editing. To answer this question, we probe the editability of pretrained flow-matching and hybrid TTS models by constructing a controlled test set and systematically diagnosing editing effects along the generative trajectory. Our analysis reveals that pretrained TTS models are substantially editable in emotion, but such editability is architecture- and trajectory-dependent and can be disrupted by early flow-matching steps, while cross-speaker emotion transport carries additional acoustic attributes beyond emotion. To address these limitations, we propose SEmoEdit, the first training-free framework that formulates emotion editing as dynamic velocity transport between source and target emotions, enabling robust, flow-based speech emotion editing directly within pretrained TTS models. SEmoEdit unifies three core operations: emotion replacement, emotion erasure, and continuous emotion interpolation, requiring neither parameter updates nor task-specific optimization. To systematically evaluate these capabilities, we introduce SEmoEditBench, a dataset comprising 600 editing cases, and conduct extensive experiments across state-of-the-art (SOTA) models and backbones. Our results show that SEmoEdit is highly effective and broadly applicable, outperforming existing training-based and activation-steering methods. Ultimately, this work reveals that pretrained speech flows possess rich, latent emotion-editing capabilities, providing useful guidance for real applications. Code, benchmark, and Audio samples are available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.34648 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T20:29:03.118Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).