Metraipolis › archive › dc_c0cb04fd334c
Frame-Wise Streaming for Efficient Learned Speech Compression — arXiv:2610.21998
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.21998
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T10:56:20.021Z
- SHA-256 of the archived text
dea64d4ee7427b726f23a186d319b5d2b46a08c9d88000e5615086a0ca6482f1- Stable URL of this page
https://metraipolis.com/documento/dc_c0cb04fd334c
How to cite
Frame-Wise Streaming for Efficient Learned Speech Compression — arXiv:2610.21998 — arXiv (en), https://arxiv.org/abs/2610.21998 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c0cb04fd334c (Metraipolis archive dc_c0cb04fd334c, sha256 dea64d4ee7427b726f23a186d319b5d2b46a08c9d88000e5615086a0ca6482f1)
BibTeX
@misc{dcc0cb04fd334c,
title = {Frame-Wise Streaming for Efficient Learned Speech Compression — arXiv:2610.21998},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.21998},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c0cb04fd334c; sha256 dea64d4ee7427b726f23a186d319b5d2b46a08c9d88000e5615086a0ca6482f1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21998. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Speech compression is increasingly important in scenarios such as real-time and satellite communication. Neural codecs have shown superior Rate-Distortion (RD) performance over traditional signal-processing codecs. Models with long-range dependency encoders/decoders and entropy models outperform convolution-based structures, but they still lack streaming support since they rely on sequence-level correlations. In this paper, we propose StreamLSC, a frame-level streaming learned speech compression model built on three fundamental components of streaming networks, i.e. Boundary-Cached Causal Convolution, Intra-Frame Layer Normalization, and RNN-mode RWKV. Our model incorporates a streamable entropy quantizer module, ensuring efficient context model while preserving causal constraints. Experiments show that the streaming model achieves performance comparable to its offline counterpart, attaining state-of-the-art rate-distortion results, significant gains in subjective evaluations, -51.37% (ViSQOL) and -53.99% (PESQ) BD-RATE savings on average, and improvements of 0.2154 and 0.5156 in BD-ViSQOL and BD-PESQ. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv eess.AS, nuovi lavori del giorno). Categorie arXiv: Sound (cs.SD) ; Audio and Speech Processing (eess.AS). Pagina del lavoro: https://arxiv.org/abs/2610.21998 — 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 2026-10-16T09:02:22.213Z dalla pagina https://arxiv.org/list/eess.AS/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).