小模型专注非洲19种语言语音识别，效果远超大模型。
WAXAL-NET: Finetuned Edge ASR Across 19 African Languages
- 用微调的小型边缘模型处理非洲方言语音。
- 平均错误率38.0%，比零样本模型低26.9个百分点。
- 适合非洲本地语音研究与轻量化部署场景。
我们评估了紧凑的领域专用语音识别模型在WAXAL语料库中19种非洲语言上的表现，是否优于大规模多语言基础模型。微调后的边缘模型取得38.0%的宏平均词错误率（WER），而最佳零样本基线为64.9%，降幅达26.9个百分点，且模型规模仅为后者的1/3至1/40。结果表明，对于非正式非洲语音，领域专化性优于模型规模。跨域测试显示，微调模型在分布外（OOD）语音上仍保持可用性能，而零样本模型在测试域与预训练分布一致时占优。通过所有语言的母语者分布式审计，构建了语言学基础的错误分类体系，揭示CTC与自回归架构在不同语系中行为差异。此外，仅看WER会低估使用音节文字语言的表现，字符错误率（CER）与WER比率显示字符级准确率显著高于表面指标。为促进未来非洲语音识别研究，我们公开所有模型权重、微调与评估脚本，以及涵盖全部19种语言的清洗版WAXAL子集。
原文摘要 · Abstract (English)
We evaluate whether compact domain-specialized ASR models can outperform massively multilingual foundation models for conversational African speech across 19 languages in the WAXAL corpus. Fine-tuned edge models achieve a macro-averaged WER of $38.0\%$ compared to $64.9\%$ for the best zero-shot baseline, a $26.9$ percentage-point reduction using models $3-40\times$ smaller. Results confirm that domain specialization dominates scale for spontaneous African speech. Cross-domain evaluation shows that fine-tuned models recover usable performance on out-of-distribution (OOD) speech, while zero-shot models regain an advantage when the test domain matches their pretraining distribution. A distributed native-speaker audit across all surveyed languages produces a linguistically-grounded error taxonomy, showing that CTC and autoregressive architectures behave differently across language families. We further show that WER alone misrepresents performance for syllabary-script languages where CER/WER ratios reveal substantially higher character-level accuracy than headline WER suggests. Finally, to contribute to future African ASR research, we release all model weights, fine-tuning and evaluation scripts, and a cleaned WAXAL subset covering all $19$ languages.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。