Metraipolis › archive › dc_7594308745bc

AfriSwitch: A Benchmark for In-the-Wild African Code-Switched Speech Recognition — arXiv:2608.26434

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.26434
License (as archived)
CC0-1.0
Archived
2026-10-17T15:19:22.792Z
SHA-256 of the archived text
3862c1e2aee642fbe928d9ae5b090a770672c0c4e1848baf3e6bb25e783c094e
Stable URL of this page
https://metraipolis.com/documento/dc_7594308745bc

How to cite

AfriSwitch: A Benchmark for In-the-Wild African Code-Switched Speech Recognition — arXiv:2608.26434 — arXiv (en), https://arxiv.org/abs/2608.26434 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7594308745bc (Metraipolis archive dc_7594308745bc, sha256 3862c1e2aee642fbe928d9ae5b090a770672c0c4e1848baf3e6bb25e783c094e)
BibTeX
@misc{dc7594308745bc,
  title = {AfriSwitch: A Benchmark for In-the-Wild African Code-Switched Speech Recognition — arXiv:2608.26434},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26434},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7594308745bc; sha256 3862c1e2aee642fbe928d9ae5b090a770672c0c4e1848baf3e6bb25e783c094e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26434. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Code-switching is pervasive in bilingual African conversation, yet most ASR systems assume monolingual input and are evaluated on curated monolingual benchmarks. We present AfriSwitch, a 61.36-hour human-transcribed benchmark of in-the-wild code-switched speech spanning 16 African languages and language varieties, released with switch-level English span tags, perutterance Code-Mixing Index (CMI), and switch-point counts. Corpus statistics show that mixing behaviour varies widely across African languages along two largely independent axes: how often speakers alternate, and how balanced the mixture is. No single scalar captures how code-switched a language is. Benchmarking five open and commercial multilingual ASR systems zero-shot yields word error rates far above published monolingual figures for the same languages, with the best system averaging 35.93% WER and no system falling below 24% on any language. Africa-targeted training, not model scale or nominal language coverage, best predicts performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.26434 — 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 2026-10-17T15:15:18.052Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).