Research Article
Smart ATM Access System using Face and Voice Recognition with Machine Learning
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357930, author={M. Rajasekaran and Machendra Machendra and S. Harish Karthick and S. Dhanush and B. Navaneetha Krishna}, title={Smart ATM Access System using Face and Voice Recognition with Machine Learning}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={face recognition voice recognition biometric authentication atm security convolutional neural networks (cnn) long short-term memory (lstm) opencv multimodal biometrics}, doi={10.4108/eai.28-4-2025.2357930} }- M. Rajasekaran
Machendra Machendra
S. Harish Karthick
S. Dhanush
B. Navaneetha Krishna
Year: 2025
Smart ATM Access System using Face and Voice Recognition with Machine Learning
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357930
Abstract
In this paper, we propose a new Automated Teller Machine (ATM) system using face and speech recognition technologies which are very effective in terms of the security and user convenience, respectively. Written in Python, this system is capable of harnessing powerful machine-learning techniques to recognise a face and authenticate a voice. This two-factor verification method guarantee that it is only verified users who are able to access their accounts and, in turn, reducing the risk of fraud and unauthorised staff access considerably. The camera of the facial recognition part records the image of the user, and input information and output information are exchanged through the voice recognition part to process voice commands, thereby realizing convenient and safe controls. The system has been implemented providing a friendly user interface and similar performance under different environmental conditions. The system, which uses libraries such as OpenCV (for visual processing) and Speech Recognition (for audio input), provides a more contemporary solution to enhance the ATM experience, given current burgeoning security issues across the financial sector. This biometric based ATM model can be taken as a platform for future developments in banking sector, by following the footsteps of adopting emerging biometric technologies even in day to-day banking, so that users may feel much safer while using digital banking systems.
