Metraipolis › archive › dc_ee03f58308fa

XSQ-AST: An Explainable Audio Spectrogram Transformer Framework for Localising Synthetic Speech Artifacts — arXiv:2609.24770

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.24770
License (as archived): CC0-1.0
Archived: 2026-10-09T07:08:53.534Z
SHA-256 of the archived text: 736b305c6565f25b4e4aebca04821ccd3890d413ab8e73d19fd2cc9e4e6a2e33
Stable URL of this page: https://metraipolis.com/documento/dc_ee03f58308fa

How to cite

XSQ-AST: An Explainable Audio Spectrogram Transformer Framework for Localising Synthetic Speech Artifacts — arXiv:2609.24770 — arXiv (en), https://arxiv.org/abs/2609.24770 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ee03f58308fa (Metraipolis archive dc_ee03f58308fa, sha256 736b305c6565f25b4e4aebca04821ccd3890d413ab8e73d19fd2cc9e4e6a2e33)

BibTeX

@misc{dcee03f58308fa,
  title = {XSQ-AST: An Explainable Audio Spectrogram Transformer Framework for Localising Synthetic Speech Artifacts — arXiv:2609.24770},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.24770},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ee03f58308fa; sha256 736b305c6565f25b4e4aebca04821ccd3890d413ab8e73d19fd2cc9e4e6a2e33},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.24770. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Localising artifacts in synthetic speech remains challenging, as most evaluation methods yield only global quality scores. This paper presents XSQ-AST, a framework that combines the SQ-AST speech quality model with WhisperX phoneme alignment and multiple saliency methods to produce temporally localised artifact diagnostics without model retraining. Saliency maps are projected onto continuous distributions via kernel density estimation and onto phoneme boundaries via phoneme-discretised saliency maps. A 40-participant listening test validated the framework across five perceptual dimensions. Attention Rollout, Attention Flow and an adapted GradCAM produced temporal distributions that correlated with listener highlights, with different methods best suited to different artifact types. An AUC-ROC analysis confirmed discrimination above chance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.LG).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Machine Learning (cs.LG); Sound (cs.SD); Signal Processing (eess.SP).
Pagina del lavoro: https://arxiv.org/abs/2609.24770 — 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 2026-10-09T06:58:09.664Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).