SOURCE-LINKED INTELLIGENCE

SpeakPay: Domain-Adaptive LoRA Fine-Tuning of Whisper for Low-Resource Nepali Financial Speech Recognition

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Mobile payment applications in Nepal are graphically mediated and largely inaccessible to visually impaired users. This paper presents SpeakPay, a voice-first digital wallet, and documents the central technical contribution: a controlled study of domain adaptation for low-resource financial speech recognition. We introduce NepFinSpeech-403, a 403-utterance dataset of Nepali financial voice commands (send, load, and balance operations spanning 237 unique numerals), and fine-tune Whisper large-v2 with LoRA. On the held-out test set, the domain-adapted model reduces Word Error Rate from 129.95% (

Read original source ↗ Open in workspace

recordType: paper
region: Global

Evidence & attribution

arXiv · AI, language, vision and robotics · 2026-09-01T18:07:24.000Z

First collected: 2026-09-21T06:01:56.170Z. This is not the publication date.