SpeakPay：LoRA微调Whisper实现尼泊尔语金融语音识别
原标题：SpeakPay: Domain-Adaptive LoRA Fine-Tuning of Whisper for Low-Resource Nepali Financial Speech Recognition
AI 摘要
SpeakPay项目针对尼泊尔语金融语音识别领域，构建了首个公开的NepFinSpeech-403数据集，并通过LoRA微调Whisper large-v2模型，将词错误率从129.95%降至42.58%，交易成功率从1.67%提升至33.33%。研究还发现100-300条领域特定语音即可获得大部分适应收益，并部署了公开的语音优先网页应用。
正文节选
SpeakPay: Domain-Adaptive LoRA Fine-Tuning of Whisper for Low-Resource Nepali Financial Speech Recognition Abstract 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 financi