Edit Content, Preserve Acoustics: Imperceptible Text-Based Speech Editing via Self-Consistency Rewards
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917238491054080 |
|---|---|
| author | Ren, Yong Yi, Jiangyan Tao, Jianhua Wen, Zhengqi Wang, Tao |
| author_facet | Ren, Yong Yi, Jiangyan Tao, Jianhua Wen, Zhengqi Wang, Tao |
| contents | Imperceptible text-based speech editing allows users to modify spoken content by altering the transcript. It demands that modified segments fuse seamlessly with the surrounding context. Prevalent methods operating in the acoustic space suffer from inherent content-style entanglement, leading to generation instability and boundary artifacts. In this paper, we propose a novel framework grounded in the principle of "Edit Content, Preserve Acoustics". Our approach relies on two core components: (1) Structural Foundations, which decouples editing into a stable semantic space while delegating acoustic reconstruction to a Flow Matching decoder; and (2) Perceptual Alignment, which employs a novel Self-Consistency Rewards Group Relative Policy Optimization. By leveraging a pre-trained Text-to-Speech model as an implicit critic -- complemented by strict intelligibility and duration constraints -- we effectively align the edited semantic token sequence with the original context. Empirical evaluations demonstrate that our method significantly outperforms state-of-the-art autoregressive and non-autoregressive baselines, achieving superior intelligibility, robustness, and perceptual quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00560 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Edit Content, Preserve Acoustics: Imperceptible Text-Based Speech Editing via Self-Consistency Rewards Ren, Yong Yi, Jiangyan Tao, Jianhua Wen, Zhengqi Wang, Tao Sound Audio and Speech Processing Imperceptible text-based speech editing allows users to modify spoken content by altering the transcript. It demands that modified segments fuse seamlessly with the surrounding context. Prevalent methods operating in the acoustic space suffer from inherent content-style entanglement, leading to generation instability and boundary artifacts. In this paper, we propose a novel framework grounded in the principle of "Edit Content, Preserve Acoustics". Our approach relies on two core components: (1) Structural Foundations, which decouples editing into a stable semantic space while delegating acoustic reconstruction to a Flow Matching decoder; and (2) Perceptual Alignment, which employs a novel Self-Consistency Rewards Group Relative Policy Optimization. By leveraging a pre-trained Text-to-Speech model as an implicit critic -- complemented by strict intelligibility and duration constraints -- we effectively align the edited semantic token sequence with the original context. Empirical evaluations demonstrate that our method significantly outperforms state-of-the-art autoregressive and non-autoregressive baselines, achieving superior intelligibility, robustness, and perceptual quality. |
| title | Edit Content, Preserve Acoustics: Imperceptible Text-Based Speech Editing via Self-Consistency Rewards |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2602.00560 |