Metraipolis › archive › dc_789291cea08d

Diagnose, Then Refine: A Closed-Loop TTS System with AudioLLM-Guided Correction — arXiv:2608.28970

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.28970
License (as archived)
CC0-1.0
Archived
2026-10-17T07:47:24.804Z
SHA-256 of the archived text
834e06805a5aece934c07f720a278254123f7b86fca43be6552986c69ecc2cbb
Stable URL of this page
https://metraipolis.com/documento/dc_789291cea08d

How to cite

Diagnose, Then Refine: A Closed-Loop TTS System with AudioLLM-Guided Correction — arXiv:2608.28970 — arXiv (en), https://arxiv.org/abs/2608.28970 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_789291cea08d (Metraipolis archive dc_789291cea08d, sha256 834e06805a5aece934c07f720a278254123f7b86fca43be6552986c69ecc2cbb)
BibTeX
@misc{dc789291cea08d,
  title = {Diagnose, Then Refine: A Closed-Loop TTS System with AudioLLM-Guided Correction — arXiv:2608.28970},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.28970},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_789291cea08d; sha256 834e06805a5aece934c07f720a278254123f7b86fca43be6552986c69ecc2cbb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28970. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Current TTS systems typically rely on open-loop, single-pass generation and can produce sporadic local prosodic defects, such as misplaced stress, unnatural pauses, or flattened intonation, that utterance-level metrics often fail to expose. We present LoopTTS, a judge-guided Filter-Judge-Refiner framework for recovering low-quality TTS outputs diagnosed by an AudioLLM. Given an initial utterance from a base TTS model, an AudioLLM Judge identifies salient prosodic issues and generates structured refine instructions; a Refiner, our fine-grained instruction-following TTS model, then performs guided expressive re-synthesis conditioned on the initial utterance, target text, and instruction. To train the Refiner, we construct Refiner-DB, a 42K-example AudioLLM-annotated dataset with word-level prosodic weak supervision. Human evaluation on diagnosed low-quality utterances shows that LoopTTS can detect perceptually salient errors and correct them with the Refiner, outperforming raw generated audio and practical open-loop re-generation baselines in recovery quality. The Refiner also demonstrates stronger instruction-following ability for stress and pause control in targeted prosody modification.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.28970 — 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 2026-10-17T07:28:26.706Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).