Metraipolis › archive › dc_32df8fdb0950

Tokens Change, Structure Endures: Spectral Watermarking for Generated Speech — arXiv:2609.33774

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33774
License (as archived)
CC0-1.0
Archived
2026-10-07T21:54:05.890Z
SHA-256 of the archived text
6f441cd172726c9159b29750e555b4b2270308cedffa2cdf87aab546310f3060
Stable URL of this page
https://metraipolis.com/documento/dc_32df8fdb0950

How to cite

Tokens Change, Structure Endures: Spectral Watermarking for Generated Speech — arXiv:2609.33774 — arXiv (en), https://arxiv.org/abs/2609.33774 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_32df8fdb0950 (Metraipolis archive dc_32df8fdb0950, sha256 6f441cd172726c9159b29750e555b4b2270308cedffa2cdf87aab546310f3060)
BibTeX
@misc{dc32df8fdb0950,
  title = {Tokens Change, Structure Endures: Spectral Watermarking for Generated Speech — arXiv:2609.33774},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33774},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_32df8fdb0950; sha256 6f441cd172726c9159b29750e555b4b2270308cedffa2cdf87aab546310f3060},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33774. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Watermarking is a promising tool for establishing the provenance of AI-generated speech. While many neural audio watermarking methods rely on a separately trained watermark generator, token-level watermarking is a training-free alternative that operates directly during generation. Its main weakness is retokenization: decoding generated speech to a waveform and encoding it again can change token identities and erode the watermark. To make the watermark robust to these changes, we propose Redwing, REtokenization-Durable Watermarking IN Generation. It builds a graph from the token substitutions observed under retokenization, whose Laplacian yields a basis that assigns similar values to tokens likely to substitute for one another. Over this basis, embedding and detection functions are jointly optimized to preserve watermark signal through retokenization while limiting embedding distortion and detector variability on unwatermarked speech. On the Moshi full-duplex system, after eight consecutive passes of Mimi resynthesis, Redwing achieves 80.7% TPR at a calibrated 1% FPR, compared with 8.3% for KGW and at most 7.3% for WMAR. It also has the highest TPR after eight passes through three other neural codecs (77.5-93.0%), and the gains generalize to TTS models at a speech-quality cost close to that of KGW. These results show that retokenization is not merely a source of noise: its transition structure can be exploited as a design principle for robust token-level watermarking.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL).
Categorie arXiv: Sound (cs.SD) ; Computation and Language (cs.CL); Cryptography and Security (cs.CR); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.33774 — 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 2026-10-07T21:27:08.779Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).