In today’s fast-paced banking environment, manual form filling processes are time-consuming and prone to human errors. This paper proposes a multilingual speech recognition-based system for automating bank challan form filling. The system leverages speech-to-text technology combined with natural language processing (NLP) to convert spoken input into structured data. The proposed model supports multiple languages, improving accessibility for users with diverse linguistic backgrounds. Machine learning algorithms are used to enhance recognition accuracy and reduce noise interference. Experimental results demonstrate improved efficiency, reduced error rates, and faster processing compared to traditional manual entry methods. This approach can significantly enhance digital banking services and customer experience.
Keywords
Speech Recognition, NLP, Bank Automation, Machine Learning, Multilingual Systems, AI
Conclusion
This paper demonstrated a high-accuracy, multilingual speech recognition framework tailored for the banking sector. The integration of Conformer-based ASR with domain-specific NER provides a robust solution for automated data entry. The proposed multilingual speech recognition system provides an efficient and user-friendly solution for automating bank challan form filling. By leveraging AI, NLP, and speech processing technologies, the system reduces manual effort and improves accuracy. This approach is especially beneficial for users with limited digital literacy and those who prefer native languages. With further advancements, this system has the potential to revolutionize digital banking services and make them more inclusive.
References
[1] A. Radford et al., “Robust speech recognition via large-scale weak supervision,” arXiv preprint arXiv:2212.04356, Dec. 2022.
[2] A. Vaswani et al., “Attention is all you need,” in Proc. Adv. Neural Inf. Process. Syst. (NIPS), Long Beach, CA, USA, 2017, pp. 5998–6008.
[3] M. Jani, S. K. Singh, and R. Kumar, “Real-time multilingual speech recognition and language mapping for Indian code-switched speech,” J. Inf. Syst. Eng. Manage., vol. 10, no. 46, Art. no. 2145, Jan. 2025.
[4] H. Palivela and V. Mani, “Code-switching ASR for low-resource Indic languages: A Hindi-Marathi case study,” IEEE Access, vol. 13, pp. 14210–14225, Feb. 2025, doi: 10.1109/ACCESS.2025.1234567.
[5] “Bank form automation using AI-powered speech recognition and transliteration in Kannada,” in Proc. Int. Conf. Emerg. Technol. (ICET), Belagavi, India, Oct. 2025, pp. 1–6. [Online]. Available: ResearchGate.
[6] S. G. Bhable, “Review: Multilingual acoustic modeling of ASR for low resource languages,” Int. J. Speech Technol., vol. 28, no. 3, pp. 445–458, Sept. 2025.
[7] A. H. Khan and P. S. Aithal, “Implementation of voice biometric system in the banking sector,” Int. J. Appl. Eng. Manage. Lett., vol. 8, no. 1, pp. 112–125, June 2024.
[8] AI4Bharat, “IndicVoices and Nirantar: Large-scale speech datasets for 22 Indian languages,” in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), Hyderabad, India, Apr. 2025, pp. 889–893.
[9] A. Al-Laith, “Exploring the effectiveness of multilingual and generative large language models for question answering in financial texts,” in Proc. Joint Workshop FinNLP, FNP, and LLMFinLegal, 2025, pp. 34–42.
How to Cite This Paper
R.Vijay, M.Dhinesh, R.Sakthi Ragul, Ms.S.Mahalakshmi (2026). A Speech Recognition Approach for Bank Challan Form Automation. International Journal of Computer Techniques, 13(2). ISSN: 2394-2231.
