Metraipolis › archive › dc_ebf37ad5a9dd
Text-Only Domain Adaptation for Low-Resource Medical ASR: Shallow Fusion and Density Ratio Rescoring for Hausa — arXiv:2611.05744
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.05744
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-05T18:17:30.104Z
- SHA-256 of the archived text
8d5463ef5c9c6f0f8873376c0dbbd8dd9e898084ecd464dad4dbdbd4169a81b4- Stable URL of this page
https://metraipolis.com/documento/dc_ebf37ad5a9dd
How to cite
Text-Only Domain Adaptation for Low-Resource Medical ASR: Shallow Fusion and Density Ratio Rescoring for Hausa — arXiv:2611.05744 — arXiv (en), https://arxiv.org/abs/2611.05744 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ebf37ad5a9dd (Metraipolis archive dc_ebf37ad5a9dd, sha256 8d5463ef5c9c6f0f8873376c0dbbd8dd9e898084ecd464dad4dbdbd4169a81b4)
BibTeX
@misc{dcebf37ad5a9dd,
title = {Text-Only Domain Adaptation for Low-Resource Medical ASR: Shallow Fusion and Density Ratio Rescoring for Hausa — arXiv:2611.05744},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.05744},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ebf37ad5a9dd; sha256 8d5463ef5c9c6f0f8873376c0dbbd8dd9e898084ecd464dad4dbdbd4169a81b4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05744. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Automatic speech recognition (ASR) systems often struggle with specialised medical vocabulary, especially in low-resource languages where domain-specific speech data are scarce. We investigate text-only domain adaptation for Hausa medical ASR using the fixed MMS-1B-FL102 acoustic model and two decoding-time language-model fusion methods: Shallow Fusion (SF) and the Density Ratio Approach (DRA). A 4-gram KenLM trained on 19,421 Hausa health sentences serves as the medical target model, while a 75,372-sentence general-domain model provides the source prior for DRA. Evaluation uses 250 recordings of 25 physician-reviewed prompts from 10 speakers across six northern Nigerian states. Under prompt-disjoint five-fold cross-validation, SF and DRA reduce WER from 29.56% to 23.55% and 23.33%, respectively. Both significantly outperform greedy decoding (p < 0.001), while their difference is not significant (p = 0.423). However, general-LM-only SF achieves 23.96% WER, indicating that the medical LM's additional improvement over a general LM is modest. Medical Term Error Rate also improves under fusion, but does not establish a distinct medical-domain advantage. These findings demonstrate that lightweight, text-only language-model fusion can improve Hausa medical ASR without additional domain-specific speech collection. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2611.05744 — 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 2026-11-05T18:12:49.032Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).