面向加纳三语青少年健康语音识别的基准测试与领域适配研究
Benchmarking and Domain Adaptation of Automatic Speech Recognition (ASR) for Adolescent Health Communication in Ghanaian Languages
加纳团队用 Qwen3-ASR-0.6B 给三种本地语言做语音识别，Ewe 错误率砍掉近一半，还上线了健康问答应用 KasaHealth，低资源语言落地的完整参考。
论文对 Twi、Dagbani、Ewe 三种加纳语言的语音识别做了端到端研究，先在圣经语料和青少年性生殖健康（ASRH）数据集上用 CER/WER 基准测试五套系统，包括三个 Wav2Vec2 模型和 Gemma 3n、Gemma 4。因 Gemma 4 微调算力不可行，研究转向 Qwen3-ASR-0.6B，用约 9 万条加纳圣经语料微调，Ewe 的 WER 从 109.3% 降到 64.8%，CER 从 65.1% 降到 24.9%。落地应用 KasaHealth 由 50 名社区受访者测试，拿到 100% 聊天通过率和 92% 推荐意愿。结论是这三种语言的瓶颈在于经过验证的领域内数据，而非模型能力或算力。
Benchmarking and Domain Adaptation of Automatic Speech Recognition (ASR) for Adolescent Health Communication in Ghanaian Languages
This paper presents an end-to-end study of automatic speech recognition (ASR) for adolescent health communication in three Ghanaian languages (Twi, Dagbani, and Ewe). The work proceeds in three connected stages; First, we benchmark five ASR systems (three language-specific Wav2Vec2 models and two multimodal LLMs, Gemma 3n and Gemma 4) on a general-domain Bible corpus and a Youth Adolescent Sexual and Reproductive Health (ASRH) Domain ASR dataset, using Character and Word Error Rate (CER, WER). Second, guided by the benchmark, we perform supervised domain adaptation: although Gemma 4 was the strongest zero-shot candidate, fine-tuning it proved computationally infeasible, so we pivoted to the compact Qwen3-ASR-0.6B, fine-tuned on a large Ghana Bible corpus (~90k samples) and evaluated strictly on held-out human-collected in-domain audio. Fine-tuning reduced WER on every language, most dramatically for Ewe (WER from 109.3% to 64.8%, a drop of 44.5 pp; CER from 65.1% to 24.9%). Third, we validate the work through KasaHealth, a live voice-first ASRH application deployed in all three languages, complemented by Senti-Check, a technical evaluation harness. KasaHealth was tested by 50 community respondents and achieved a 100% chat-approval rate, a 72% Good-or-Excellent translation rating, and a 92% would-recommend rate, while surfacing the domain gaps that most constrain real-world use. Across all three stages the evidence converges: for these languages the binding constraint is validated in-domain data, not model capability or computation.