Voice-based user interface for hands-free data entry and automation at workplaces

IF 1.9 Q2 MULTIDISCIPLINARY SCIENCES

MethodsX Pub Date : 2025-08-28 DOI:10.1016/j.mex.2025.103596

Daiwiik Harihar, Vedansh Shrivastava, Pratvina Talele, Aditi Jahagirdar

{"title":"Voice-based user interface for hands-free data entry and automation at workplaces","authors":"Daiwiik Harihar, Vedansh Shrivastava, Pratvina Talele, Aditi Jahagirdar","doi":"10.1016/j.mex.2025.103596","DOIUrl":null,"url":null,"abstract":"<div><div>The increasing demand for hands-free interaction in modern workplaces has led to the development of Voice-Based User Interfaces (VUIs) that enhance accessibility, efficiency and automation. This research presents the Voice-Based User Interface for Hands-Free Data Entry and Automation at Workplaces. The system enables real-time speech-to-text transcription, allowing users to interact with workplace applications without manual input, making it intuitive, user-friendly and capable of enhancing efficiency and convenience in various workplace scenarios. Through extensive testing and evaluation, the study demonstrates the practicality and benefits of the Voice-Based User Interface for hands-free data entry and automation.</div><div><strong>Methodology Overview</strong>:<ul><li><span>•</span><span><div>Utilized WIT.AI API for speech-to-text transcription.</div></span></li><li><span>•</span><span><div>Implemented chunking, caching, and concurrency control to optimize processing.</div></span></li><li><span>•</span><span><div>Evaluated performance using Word Error Rate (WER), Levenshtein Distance and Cosine Similarity on real world datasets.</div></span></li></ul>The system proves to be upto 88.8% accurate in recognizing spoken commands and efficiently converting them into text with best performance achieved when the audio was divided into 7 optimal chunks. Cosine Similarity for these chunks is more accurate than that of sizeable file and approximately 2. Moreover, the integration of real-time updates across different domains (educational, legal, medical) and data synchronization enhances productivity and usability. In conclusion, the Voice-Based User Interface offers a viable solution for hands-free data entry and automation at workplaces.</div></div>","PeriodicalId":18446,"journal":{"name":"MethodsX","volume":"15 ","pages":"Article 103596"},"PeriodicalIF":1.9000,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"MethodsX","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2215016125004406","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MULTIDISCIPLINARY SCIENCES","Score":null,"Total":0}

引用次数: 0

Abstract

The increasing demand for hands-free interaction in modern workplaces has led to the development of Voice-Based User Interfaces (VUIs) that enhance accessibility, efficiency and automation. This research presents the Voice-Based User Interface for Hands-Free Data Entry and Automation at Workplaces. The system enables real-time speech-to-text transcription, allowing users to interact with workplace applications without manual input, making it intuitive, user-friendly and capable of enhancing efficiency and convenience in various workplace scenarios. Through extensive testing and evaluation, the study demonstrates the practicality and benefits of the Voice-Based User Interface for hands-free data entry and automation.

Methodology Overview:

•
Utilized WIT.AI API for speech-to-text transcription.
•
Implemented chunking, caching, and concurrency control to optimize processing.
•
Evaluated performance using Word Error Rate (WER), Levenshtein Distance and Cosine Similarity on real world datasets.

The system proves to be upto 88.8% accurate in recognizing spoken commands and efficiently converting them into text with best performance achieved when the audio was divided into 7 optimal chunks. Cosine Similarity for these chunks is more accurate than that of sizeable file and approximately 2. Moreover, the integration of real-time updates across different domains (educational, legal, medical) and data synchronization enhances productivity and usability. In conclusion, the Voice-Based User Interface offers a viable solution for hands-free data entry and automation at workplaces.

Abstract Image

查看原文本刊更多论文

基于语音的用户界面，用于工作场所的免提数据输入和自动化

现代工作场所对免提交互的需求日益增长，导致了基于语音的用户界面（VUIs）的发展，以提高可访问性、效率和自动化。本研究提出了基于语音的用户界面，用于工作场所的免提数据输入和自动化。该系统可实现实时语音转文本转录，允许用户无需手动输入即可与工作场所应用程序交互，使其直观，用户友好，能够在各种工作场所场景中提高效率和便利性。通过广泛的测试和评估，该研究证明了基于语音的用户界面在免提数据输入和自动化方面的实用性和优势。方法概述：•利用智慧。用于语音到文本转录的AI API。•实现分块，缓存和并发控制，以优化处理。•使用单词错误率（WER）， Levenshtein距离和余弦相似度在真实世界数据集上评估性能。事实证明，该系统在识别语音命令并有效地将其转换为文本方面的准确率高达88.8%，当音频被分成7个最佳块时，性能达到最佳。这些块的余弦相似度比大文件更准确，大约为2。此外，跨不同领域（教育、法律、医疗）的实时更新和数据同步的集成提高了生产力和可用性。总之，基于语音的用户界面为工作场所的免提数据输入和自动化提供了一个可行的解决方案。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊