Authors :
Ragavendran H.
Volume/Issue :
Volume 10 - 2025, Issue 10 - October
Google Scholar :
https://tinyurl.com/27ymup5n
Scribd :
https://tinyurl.com/2ks4h6j6
DOI :
https://doi.org/10.38124/ijisrt/25oct239
Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.
Note : Google Scholar may take 30 to 40 days to display the article.
Abstract :
Modern eCommerce sites grapple with outdated search systems that use keywords to perform searches and
provide a non-smooth product search experience, as well as restrict access to a wider range of user bases. In this study,
Speak Search Shop is introduced, a voice-based product search model with React.JS as a frontend, Web Speech API as an
orchestration microservice, and Elasticsearch as a contextual search microservice. The specification of requirements takes
place in a form of a discussion; the user requests running shoes at a price below seven thousand rupees with an insole; these
are then classified by intent and extracted into structured search terms. With 5,000 voice queries, experimental validation
produced 94.2% intent classification accuracy, 91.7% entity extraction F1-score and average 687-milliseconds. Comparative
analysis revealed enhanced first-result relevance by 37 percent, reduced zero-result queries by 42 percent, and reduced time
to complete the task by 41 percent compared with text-based search. The 150-user test revealed a 33 percent increase in
conversion rate and a significant boost in satisfaction, especially among the visually impaired users. The microservice
containerized architecture embraces horizontal, Kube-based scaling, and can adjust to India English accent differences and
code-switching habits. This architecture confirms the commercial viability of conversational interfaces as alternatives to
traditional search, and shows that transformer-based language models can be integrated with production-scale
infrastructure. Directions Future directions Multimodal search Multimodal search integrates voice and visual search,
regional language support, and customized recommendation. The study brings reference designs of voice commerce systems
to fit actual scale, latency, and precision requirements and improve accessibility in Internet retail settings.
Keywords :
Interactive Voice Search E-Commerce Platforms, Natural Language Processing (NLP), Large Language Models (LLM), React.js Frontend, Python, Node.js Microservices, Elastic Search, Human-Computer Interaction.
References :
- Kandhari, M. S., Zulkemine, F., & Isah, H. (2018, November). A voice-controlled e-commerce web application. In 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON) (pp. 118-124). IEEE.
- Duttaroy, N., Angre, A., Powar, S., Patil, P., & Hegde, G. (2022). Voice controlled E-commerce web app. International Research Journal of Engineering and Technology (IRJET), 1(9).
- Bhalla, A., Garg, S., & Singh, P. (2020). Present day web-development using reactjs. International Research Journal of Engineering and Technology, 7(05).
- Sayali Sunil Tandel, Abhishek Jamadar, “Impact of Progressive Web Apps on Web App Development,”International Journal of Innovative Research in Science,Engineering and Technology [IJIRSET] Vol. 7, Issue 9, September 2018.
- M. S. Kandhari, F. Zulkemine and H. Isah, "A Voice Controlled E-Commerce Web Application," 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), Vancouver, BC, Canada, 2018, pp. 118- 124, doi:10.1109/IEMCON.2018.8614771.
- Kunal Mohadikar and Rahul Nawkhare, "Ecommerce based online shopping for visually impaired people using speech recognition", International Journal of Development Research Vol. 07, Issue, 08, pp.14581-14584, August, 2017.
- Shraddha Londhe and Dr. Hemant Deshmukh, "Review Paper on Search-Engine Optimization", International Journal of Engineering Science and Computing (IJESC) Vol. 07, Issue, 04, 2017.
- Momin Mukherjee, Sahadev Roy,”E-Commerce and Online Payment in the Modern Era”, International Journal of Advanced Research in Computer Science and Software Engineering.,Volume 7, Issue 5, May 2017
- Shravan G V, Anitha Sandeep, "Comprehensive Analysis for React-Redux Development Framework", International Journal of Creative Research Thoughts (IJCRT)Volume: 08 Issue: 04, April 2020.
- Amarnath Vishwakarma, Suchi Sharma, Esha Bhardwaj, “E-Commerce Site and Its Development”, International Research Journal of Engineering and Technology (IRJET), ENCADEMS – 2017 (Volume 4 – Issue 03)
- Archana Bhalla, Shivangi Garg, Priyangi Singh, “Present Day Web-Development Using ReactJs”, International Research Journal of Engineering and Technology (IRJET), Volume 7 – Issue 05, May 2020.
- Mohammad Ahmad, Mohammad Faris, Patel Suhel I,”Search Engine Optimization (SEO)”, International Research Journal of Engineering and Technology (IRJET), Volume: 06 Issue: 04, Apr 2019.
- Ritwik C and Anitha Sandeep, “ReactJs and Front-End Development”, International Research Journal of Engineering and Technology (IRJET), Volume: 07 Issue: 04, Apr 2020.
- Rahul Surendra Mishra, "Progressive WEBAPP: Review", International Journal of Development Research, Volume: 03 Issue: 06, June-2016.
- Ashutosh Tiwari, Smriti Srivastava, “Exploring Front End Framework React: A review”, International Research Journal of Engineering and Technology (IRJET), Volume: 07 Issue: 07, July 2020.
- S. Sagar, V. Awasthi, S. Rastogi, T. Garg, S. Kuzhalvaimozhi, “Web application for voice operated e-mail exchange”,
- V. K{\"e}puska and B. Gamal, “Comparing speech recognition systems (Microsoft API, Google API and CMU Sphinx,” Journal of Engineering Research and Application, vol. 7, no. 3, pp. 20-24, 2017.
- C. Gaida, P. Lange, R. Petrick, P. Proba, A. Malatawy and D. Suendermann-Oeft, “Comparing open-source speech recognition toolkits,” Tech. Rep., DHBW Stuttgart, 2014.
- Gazetić, “Comparison Between Cloud-based and Offline Speech Recognition Systems”. Master’s thesis. Technical University of Munich, Munich, Germany, 2017.
- M. Alshamari, “Accessibility evaluation of Arabic e-commerce web sites using automated tools,” Journal of Software Engineering and Applications, vol. 9, no. 09, p. 439, 2016.
- C. McNair, “Worldwide retail and ecommerce sales: emarketer’s updated forecast and new mcommerce estimates for 2016–2021,” Industry Report, eMarketing, 2018.
- Y. Borodin, J. P. Bigham, G. Dausch and I. Ramakrishnan, “More than meets the eye: a survey of screen-reader browsing strategies,” Proceedings of the 2010 International Cross Disciplinary Conference on Web Accessibility (W4A), p. 13, 2010.
- Alison DeNisco Rayome, 2017, “Why IBM's speech recognition breakthrough matters for AI and IoT”, white paper, online at: https://www.techrepublic.com/article/why-ibms-speech-recognition-breakthrough-matters-for-ai-and-iot/.
Modern eCommerce sites grapple with outdated search systems that use keywords to perform searches and
provide a non-smooth product search experience, as well as restrict access to a wider range of user bases. In this study,
Speak Search Shop is introduced, a voice-based product search model with React.JS as a frontend, Web Speech API as an
orchestration microservice, and Elasticsearch as a contextual search microservice. The specification of requirements takes
place in a form of a discussion; the user requests running shoes at a price below seven thousand rupees with an insole; these
are then classified by intent and extracted into structured search terms. With 5,000 voice queries, experimental validation
produced 94.2% intent classification accuracy, 91.7% entity extraction F1-score and average 687-milliseconds. Comparative
analysis revealed enhanced first-result relevance by 37 percent, reduced zero-result queries by 42 percent, and reduced time
to complete the task by 41 percent compared with text-based search. The 150-user test revealed a 33 percent increase in
conversion rate and a significant boost in satisfaction, especially among the visually impaired users. The microservice
containerized architecture embraces horizontal, Kube-based scaling, and can adjust to India English accent differences and
code-switching habits. This architecture confirms the commercial viability of conversational interfaces as alternatives to
traditional search, and shows that transformer-based language models can be integrated with production-scale
infrastructure. Directions Future directions Multimodal search Multimodal search integrates voice and visual search,
regional language support, and customized recommendation. The study brings reference designs of voice commerce systems
to fit actual scale, latency, and precision requirements and improve accessibility in Internet retail settings.
Keywords :
Interactive Voice Search E-Commerce Platforms, Natural Language Processing (NLP), Large Language Models (LLM), React.js Frontend, Python, Node.js Microservices, Elastic Search, Human-Computer Interaction.