aiqbench analysis
This study demonstrates that for low-resource African languages, the primary bottleneck in deploying speech recognition for health communication is the scarcity of validated in-domain data rather than raw model capability. By pivoting from computationally heavy zero-shot LLMs to a compact, fine-tuned open-source model, researchers achieved significant error rate reductions and successful real-world deployment in adolescent health applications. The findings underscore that practical AI adoption in underserved regions depends on curating specific datasets to bridge the gap between general pre-training and domain-specific utility.
Benchmarking and Domain Adaptation of Automatic Speech Recognition (ASR) for Adolescent Health Communication in Ghanaian Languages
arXiv:2609.29798v1 Announce Type: new Abstract: 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.
Abstract: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.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
I.2.7 [Artificial Intelligence]: Natural Language Processing, J.3 [Computer Applications]: Life and Medical Sciences, I.5.4 [Pattern Recognition]: Applications
arXiv:2609.29798 [cs.CL]
(or arXiv:2609.29798v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.29798 arXiv-issued DOI via DataCite (pending registration)
Full text under the publisher's open licence · 264 words
Related coverage
Evaluate Clinical ASR Models Faster with Agent Skills and NVIDIA Nemotron Speech
NVIDIA · Jun 9, 2026
Reduce ASR inference costs by 75% with NVIDIA MPS on Amazon EC2
AWS · Aug 27, 2026
Curriculum-Based Noise Adaptation for Phoneme-to-Text Reconstruction in Visual Speech Recognition
arXiv · Sep 21, 2026
How Robust Are Neural Audio Codecs for African Speech? A Multi-Task Benchmark and the Limits of Perceptual Quality
arXiv · Oct 2, 2026