Metraipolis › archive › dc_cef29ca87b7f
Language-Statistical Analysis of Neural Audio Codec Tokens Across Architectures, Corpora, and Noise Conditions — arXiv:2608.31037
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.31037
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T08:07:52.904Z
- SHA-256 of the archived text
c009fe8c52b246074a1dc29cf820fa8366c7662c1d344d9a29d823d435c4b8f2- Stable URL of this page
https://metraipolis.com/documento/dc_cef29ca87b7f
How to cite
Language-Statistical Analysis of Neural Audio Codec Tokens Across Architectures, Corpora, and Noise Conditions — arXiv:2608.31037 — arXiv (en), https://arxiv.org/abs/2608.31037 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cef29ca87b7f (Metraipolis archive dc_cef29ca87b7f, sha256 c009fe8c52b246074a1dc29cf820fa8366c7662c1d344d9a29d823d435c4b8f2)
BibTeX
@misc{dccef29ca87b7f,
title = {Language-Statistical Analysis of Neural Audio Codec Tokens Across Architectures, Corpora, and Noise Conditions — arXiv:2608.31037},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.31037},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cef29ca87b7f; sha256 c009fe8c52b246074a1dc29cf820fa8366c7662c1d344d9a29d823d435c4b8f2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.31037. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Neural audio codecs (NACs) convert speech into discrete token sequences, and prior work has reported that these sequences follow language-like statistical laws. This paper analyzes the token statistics of 13 NACs spanning multi-codebook residual vector quantization (RVQ), single-codebook VQ, and non-VQ designs, evaluated on three corpora under clean, white-noise, and real-world DEMAND-noise conditions. Zipf and Heaps parameters, unigram entropy, codebook occupancy, and Jensen-Shannon divergence (JSD) are estimated from matched token samples with explicit fit-validity safeguards and family-conditional $n$-gram orders. Corpus identity explains little variance in any metric, whereas acoustic condition and quantizer meta-category dominate in a metric-dependent way, and unigram entropy is the metric most strongly associated with meta-category. Clean-to-noise JSD computed at a common unigram order is associated with mel-cepstral distortion most clearly under DEMAND noise. The collapse and explosion degradation signatures previously reported for RVQ codecs concentrate in RVQ cells under white and DEMAND noise, respectively; explosion also occurs in non-VQ codecs, and single-codebook VQ codecs shift in occupancy and distribution shape without either signature. These results provide architecture-conditioned conventions for applying language-statistical analysis to NAC tokens. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Sound (cs.SD). Pagina del lavoro: https://arxiv.org/abs/2608.31037 — 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 2026-10-17T07:47:52.878Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).