Metraipolis › archive › dc_dfd3eefc1981

SyRA-CS: Synthetic-to-Real Adaptation for Code-Switched ASR with Meta-Learning — arXiv:2610.31415

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.31415
License (as archived): CC0-1.0
Archived: 2026-10-25T08:36:34.394Z
SHA-256 of the archived text: e08e1179485f5b93abd951ada797ca7835251f26bfb4144eddaa2944c70075e8
Stable URL of this page: https://metraipolis.com/documento/dc_dfd3eefc1981

How to cite

SyRA-CS: Synthetic-to-Real Adaptation for Code-Switched ASR with Meta-Learning — arXiv:2610.31415 — arXiv (en), https://arxiv.org/abs/2610.31415 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dfd3eefc1981 (Metraipolis archive dc_dfd3eefc1981, sha256 e08e1179485f5b93abd951ada797ca7835251f26bfb4144eddaa2944c70075e8)

BibTeX

@misc{dcdfd3eefc1981,
  title = {SyRA-CS: Synthetic-to-Real Adaptation for Code-Switched ASR with Meta-Learning — arXiv:2610.31415},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31415},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dfd3eefc1981; sha256 e08e1179485f5b93abd951ada797ca7835251f26bfb4144eddaa2944c70075e8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31415. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Code-switched (CS) automatic speech recognition (ASR) remains difficult because real CS speech is scarce and synthetic speech does not fully match it. We investigate whether synthetic CS speech is more useful for learning an adaptation-friendly initialization than for ordinary continual fine-tuning. We propose SyRA-CS, a pair-specific synthetic-to-real adaptation framework based on Reptile meta-learning. For each language pair and pretrained model, we meta-train on the corresponding synthetic CS-FLEURS subset and then adapt using one hour of labeled real target speech. We evaluate four architectural settings: Encoder-CTC, FastConformer-RNN-T, Attention Encoder-Decoder (AED) and Speech-LLM. Experiments cover on Arabic-, French-, German-, and Hindi-English, against synthetic-only transfer, real-only fine-tuning, and continual fine-tuning. SyRA-CS improves over continual fine-tuning on WER in all 16 model-pair comparisons and on CER in 15 of 16, with relative macro WER reductions of 17.7% for XLS-R and 5.3-5.8% for the other three models. The gains hold across the tested models but vary in size across architectures, language pairs and metrics.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.31415 — 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 2026-10-25T08:31:27.803Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).