WaxalNet：19 种非洲语言自发语音识别基准测试与模型适配
WaxalNet: Benchmarking and Adapting Speech Recognition for Spontaneous Speech in 19 African Languages
WaxalNet 在覆盖 19 种非洲语言自发语音的 WAXAL 语料上评测了 Whisper、MMS、Omnilingual ASR 和 SeamlessM4T-v2 的零样本与微调表现，微调将宏平均 WER 从最佳零样本模型 Omnilingual-7B 的 52.3 降至 300M 参数 MMS 的 36.2、Omnilingual 配 LLM 解码器的 31.5。
Authors:Victor Tolulope Olufemi, Oreoluwa Babatunde, Ramsey Njema, Bolarinwa Gbotemi, Wanchi Lucia Yen, John Uzodinma, Sunday Ajayi, Oluwademilade Williams, Kausar Moshood, Innocent Elendu Anyaele, Akebert Arefaine, Candace Hunzwi, Wongel Dawit Daniel, Emmilly Namuganga, Cleophas Kadima, Athanase Bahizire, Onitsiky Ranaivoson, Emmanuel Aaron, Nicholaus Ladislaus, Idris Muhammed, Jonathan Enoch Simenya, Martin Koome, Matewos Tegete Endaylalu, Peter Ifeoluwa Adeyemo, Hondi Prisca Birindwa, Ukachi Agnes Eze-Mbey, Yacoba Oduro-Yeboah, Pericles Adjovi, Mikel K. Ngueajio, Toluwani Aremu, Prasenjit Mitra
Abstract:Multilingual speech foundation models claim broad language coverage, but how well they handle spontaneous African speech, and how best to adapt them, remains unclear. We benchmark zero-shot and fine-tuned speech recognition on the WAXAL corpus of spontaneous speech in 19 African languages, covering Whisper, MMS, Omnilingual ASR, and SeamlessM4T-v2. Fine-tuning lowers macro word error rate (WER) from 52.3 for the best zero-shot model covering all 19 languages (Omnilingual-7B) to 36.2 for a 300M-parameter MMS model and 31.5 for Omnilingual with an LLM decoder, although zero-shot models are already competitive on a few languages. Using Omnilingual's shared encoder, we compare CTC and LLM decoders with encoder pretraining held fixed. Between MMS and Whisper, the better model depends on the language family, but with a shared encoder the LLM decoder has lower WER on 18 of 19 languages, so the family pattern more likely comes from pretraining than from the decoder. The LLM decoder is, however, about five times larger and much slower on CPU. Pooling languages into one model, by family or across all 19, costs only 0.6 to 1.0 WER on average relative to per-language models, a modest price for serving all languages with one model. Fine-tuned models remain usable on read speech, and for held-out languages neither training on the other languages nor in-context examples replace in-language fine-tuning. A native-speaker audit across all 19 languages characterizes failure modes such as repetition loops. We release the code and a suite of over 100 fine-tuned models.
| Subjects: | Computation and Language (cs.CL); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC) |
| Cite as: | arXiv:2606.02375 [cs.CL] |
| (or arXiv:2606.02375v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2606.02375 arXiv-issued DOI via DataCite |
Submission history
From: Victor Tolulope Olufemi [view email]
[v1]
Mon, 1 Jun 2026 15:22:35 UTC (1,862 KB)
[v2]
Wed, 21 Oct 2026 23:38:29 UTC (45 KB)
来源：arXiv cs.CY · arxiv.org