Metraipolis › archive › dc_7ae44d6f990e

TinyAudio: Compact and Efficient Text-to-Audio Generation for Low-Resource Deployment — arXiv:2609.31525

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31525
License (as archived)
CC0-1.0
Archived
2026-10-08T04:19:24.266Z
SHA-256 of the archived text
38a8f464e21d9e33211ceb4b294e75ff465d871b6be7efb97a5fe87c385b3246
Stable URL of this page
https://metraipolis.com/documento/dc_7ae44d6f990e

How to cite

TinyAudio: Compact and Efficient Text-to-Audio Generation for Low-Resource Deployment — arXiv:2609.31525 — arXiv (en), https://arxiv.org/abs/2609.31525 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7ae44d6f990e (Metraipolis archive dc_7ae44d6f990e, sha256 38a8f464e21d9e33211ceb4b294e75ff465d871b6be7efb97a5fe87c385b3246)
BibTeX
@misc{dc7ae44d6f990e,
  title = {TinyAudio: Compact and Efficient Text-to-Audio Generation for Low-Resource Deployment — arXiv:2609.31525},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31525},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7ae44d6f990e; sha256 38a8f464e21d9e33211ceb4b294e75ff465d871b6be7efb97a5fe87c385b3246},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31525. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Text-to-audio (TTA) generation has advanced rapidly in generation quality and instruction following. However, representative systems often require around a billion parameters, limiting deployment on resource-constrained devices. This paper introduces TinyAudio, a compact flow-matching-based TTA model for low-resource deployment. At its core, TinyAudio uses TA-DiT, a 35M single-stream flow-matching Transformer. TinyAudio also includes TA-CLAP, a 32M audio-aligned text encoder, and TA-VAE, whose 20M decoder reconstructs 44.1 kHz audio from compressed latents. TinyAudio has only 87M parameters in total, over 90% fewer than representative billion-parameter pipelines, and uses 0.48 GB peak GPU memory. TinyAudio achieves competitive generation quality on AudioCaps and TTA-Bench. We further introduce TinyAudio-MF, a MeanFlow-accelerated model that enables real-time generation with a four-core CPU quota. Our results demonstrate a practical quality-footprint trade-off for low-resource TTA deployment.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in eess.AS).
Categorie arXiv: Sound (cs.SD) ; Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.31525 — 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 2026-10-08T03:50:06.048Z dalla pagina https://arxiv.org/catchup/eess.AS/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).