Research Papers 论文研究 7h ago Updated 2h ago 更新于 2小时前 44

SpeakPay: Domain-Adaptive LoRA Fine-Tuning of Whisper for Low-Resource Nepali Financial Speech Recognition SpeakPay:面向低资源尼泊尔语金融语音识别的Whisper领域自适应LoRA微调

SpeakPay introduces a voice-first digital wallet for Nepal, addressing inaccessibility of mobile payment apps for visually impaired users The authors release NepFinSpeech-403, a 403-utterance dataset of Nepali financial voice commands covering 237 unique numerals across send, load, and balance operations Fine-tuning Whisper large-v2 with LoRA reduces Word Error Rate from 129.95% to 42.58% (67.2% relative reduction) and improves Devanagari numeral recognition from 0.0% to 73.9% Task-level impact 提出SpeakPay语音优先数字钱包,解决尼泊尔视障用户移动支付可访问性问题 构建NepFinSpeech-403数据集(403条尼泊尔语金融语音指令,涵盖237个唯一数字) 使用LoRA微调Whisper large-v2,WER从129.95%降至42.58%(相对减少67.2%) 交易成功率从1.67%提升至33.33%(约20倍提升),统计显著性p < 10^-17 仅需100条领域数据即可减半WER,300条时性能趋于稳定,验证低资源场景可行性

55
Hot 热度
72
Quality 质量
65
Impact 影响力

Analysis 深度分析

TL;DR

  • SpeakPay introduces a voice-first digital wallet for Nepal, addressing inaccessibility of mobile payment apps for visually impaired users
  • The authors release NepFinSpeech-403, a 403-utterance dataset of Nepali financial voice commands covering 237 unique numerals across send, load, and balance operations
  • Fine-tuning Whisper large-v2 with LoRA reduces Word Error Rate from 129.95% to 42.58% (67.2% relative reduction) and improves Devanagari numeral recognition from 0.0% to 73.9%
  • Task-level impact far exceeds word-level metrics: Transaction Success Rate improves from 1.67% to 33.33%, a roughly 20x gain
  • As few as 100 domain-specific utterances halve the zero-shot WER, with performance plateauing around 300 examples

Why It Matters

This work demonstrates that domain-adaptive fine-tuning of large pre-trained speech models can dramatically improve performance on low-resource languages and specialized domains, making it directly relevant for practitioners working on accessibility and financial technology in underrepresented linguistic contexts. The finding that word-level metrics substantially understate practical impact is a critical reminder for the community to evaluate speech recognition systems using task-oriented metrics rather than relying solely on WER.

Technical Details

  • Model: Whisper large-v2 fine-tuned with Low-Rank Adaptation (LoRA) for domain adaptation to Nepali financial speech
  • Dataset: NepFinSpeech-403, comprising 403 utterances of Nepali financial voice commands (send, load, balance operations) spanning 237 unique numerals in Devanagari script
  • Performance: Zero-shot WER of 129.95% reduced to 42.58% after fine-tuning; Devanagari numeral recognition accuracy improved from 0.0% to 73.9%; Transaction Success Rate increased from 1.67% to 33.33%
  • Data efficiency: Performance halves with as few as 100 utterances and plateaus around 300 examples, indicating strong data efficiency of LoRA fine-tuning
  • Error analysis: Systematic numeral confusion patterns identified, including zero insertion/deletion and prefix hallucination, which account for the majority of remaining transaction failures; statistical significance confirmed via sign test (p < 10^-17)

Industry Insight

  • Low-resource language speech systems can achieve transformative practical improvements through lightweight domain adaptation (LoRA) rather than requiring massive labeled datasets, lowering the barrier for inclusive AI deployment in emerging markets
  • Researchers and practitioners should adopt task-level evaluation metrics (e.g., Transaction Success Rate) alongside traditional WER, as word-level errors can disproportionately impact real-world application performance, especially in numeracy-critical domains like finance
  • The data efficiency findings suggest that organizations can achieve meaningful speech recognition improvements with minimal domain-specific data, enabling rapid prototyping and deployment of voice-first interfaces for accessibility-focused products

TL;DR

  • 提出SpeakPay语音优先数字钱包,解决尼泊尔视障用户移动支付可访问性问题
  • 构建NepFinSpeech-403数据集(403条尼泊尔语金融语音指令,涵盖237个唯一数字)
  • 使用LoRA微调Whisper large-v2,WER从129.95%降至42.58%(相对减少67.2%)
  • 交易成功率从1.67%提升至33.33%(约20倍提升),统计显著性p < 10^-17
  • 仅需100条领域数据即可减半WER,300条时性能趋于稳定,验证低资源场景可行性

为什么值得看

本文展示了在低资源语言(尼泊尔语)和金融垂直领域的语音识别适配方案,为视障群体数字包容性提供了实用技术路径。研究揭示了词级指标与实际任务表现的巨大差距,强调领域适配在垂直场景中的关键价值。

技术解析

  • 数据集构建:NepFinSpeech-403包含403条尼泊尔语金融语音指令,涵盖发送、充值、余额查询三类操作,涉及237个唯一数字,专为低资源金融语音识别设计
  • 模型方法:基于Whisper large-v2,使用LoRA进行领域自适应微调,在保持基础模型能力的同时快速适配金融领域
  • 性能提升:WER从129.95%降至42.58%(相对减少67.2%),天城文数字识别准确率从0%提升至73.9%
  • 任务级效果:交易成功率从1.67%提升至33.33%(约20倍提升),证明领域适配的实际业务价值远超词级指标
  • 数据效率分析:100条领域数据即可减半WER,约300条时性能趋于稳定,为低资源场景提供数据规模参考
  • 错误模式分析:主要错误为零插入/删除、前缀幻觉,这些系统性错误导致大部分交易失败,为后续优化指明方向

行业启示

  • 低资源语言场景下,领域适配比单纯扩大模型规模更具性价比,100-300条高质量领域数据即可显著改善性能,为小众语言AI应用提供可行路径
  • 语音交互设计应关注任务级指标(如交易成功率)而非仅优化词级指标,两者可能存在巨大差距,需建立更贴近业务场景的评估体系
  • 金融类语音应用需特别关注数字识别准确性,系统性错误模式(如零的插入/删除)对交易安全影响显著,应在模型设计和后处理中针对性优化

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Speech 语音 Fine-tuning 微调 Dataset 数据集 Finance AI 金融AI Research 科学研究