Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{196014,
        author = {Pasadi Tanuja and Parri Praveen Ganesh and Sharmila Bhavani Polamuri and Pothala Sampath kumar},
        title = {Artificial intelligence based lie detection using facial expressions and voice tone},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {1409-1415},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196014},
        abstract = {Lie detection plays a crucial role in areas such as criminal investigations, security screening, corporate recruitment, and psychological assessment. Traditional lie detection techniques such as polygraph tests rely on physiological signals like heart rate and blood pressure, which require specialized hardware and controlled environments. However, these methods are expensive, intrusive, and sometimes unreliable.
This research proposes an AI-Based Multimodal Lie Detection System using Facial Expression Analysis and Voice Stress Analysis implemented in Python. The system integrates computer vision and audio signal processing techniques to detect deceptive behavior. Facial expressions are analyzed using deep learning-based emotion recognition models, while voice features such as pitch, frequency, and tone variation are extracted using audio processing libraries. These multimodal features are combined and classified using machine learning algorithms to determine whether a person is likely telling the truth or lying.
The proposed system provides a non-invasive, cost-effective, and automated approach to deception detection. Experimental evaluation demonstrates that combining facial and vocal cues improves prediction accuracy compared to single-modal systems.},
        keywords = {},
        month = {April},
        }