Metraipolis › archive › dc_afc3327a6774

Geometric Iterative Retrieval for Neural Audio Codec Resynthesis — arXiv:2608.19141

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.19141
License (as archived)
CC0-1.0
Archived
2026-10-18T13:53:55.476Z
SHA-256 of the archived text
cff4c56f48186c30b89dacd4e52d56c23074b73e9df977f794164d19df33005d
Stable URL of this page
https://metraipolis.com/documento/dc_afc3327a6774

How to cite

Geometric Iterative Retrieval for Neural Audio Codec Resynthesis — arXiv:2608.19141 — arXiv (en), https://arxiv.org/abs/2608.19141 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_afc3327a6774 (Metraipolis archive dc_afc3327a6774, sha256 cff4c56f48186c30b89dacd4e52d56c23074b73e9df977f794164d19df33005d)
BibTeX
@misc{dcafc3327a6774,
  title = {Geometric Iterative Retrieval for Neural Audio Codec Resynthesis — arXiv:2608.19141},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.19141},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_afc3327a6774; sha256 cff4c56f48186c30b89dacd4e52d56c23074b73e9df977f794164d19df33005d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19141. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural audio codecs based on Residual Vector Quantization (RVQ) have become the dominant discrete representation for token-based general audio generation, yet resynthesizing high-quality audio from coarse codec tokens remains an open problem and bounds the fidelity of every system that generates them. Prior work has framed resynthesis as a choice between discrete token prediction and continuous regression. We argue that this dichotomy is incomplete and introduce geometric iterative retrieval, a paradigm that uses the RVQ layer hierarchy itself as a natural iterative decomposition in continuous codebook space. Rather than classifying over discrete vocabularies or regressing to a single target vector, our method performs contrastive retrieval in the codebook's geometric space. We evaluate our method on codec restoration tasks across speech and music, and show improvements over both single-pass token prediction and one-step regression baselines.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-20 in cs.LG).
Categorie arXiv: Sound (cs.SD) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.19141 — 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 2026-10-18T13:50:59.012Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).