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{208735,
        author = {Sukhada Sandeep Alhat and Anaya Ganesh More and Mrs. Guna Dhondwad},
        title = {Deepfake Vishing 2.0: Real-Time Defenses Against AI Driven Voice Cloning Attacks},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {610-613},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=208735},
        abstract = {By 2026, real-time voice cloning combined with large language models (LLMs) will enable fully interactive vishing (voice phishing) attacks, where adversaries dynamically impersonate executives, family members, or IT support during live phone calls. Unlike traditional deepfake audio, which is pre-recorded and analyzed offline, this emerging threat demands detection that operates while the call is happening. This paper investigates the linguistic, paralinguistic, and behavioral markers such as response latency, semantic drift, and contextual knowledge gaps that distinguish an AI-generated impostor from a genuine speaker during live dialogue. It proposes lightweight, real-time defenses including streaming audio authentication and co-presence challenge-response protocols that verify identity without disrupting conversation flow. Through a controlled voice-cloning simulation and a red-teaming study with corporate employees, the research evaluates human susceptibility to interactive impersonation and compares behavioral training against real-time technical warnings, culminating in a prototype "Conversational Guardian" system for continuous in call authentication.},
        keywords = {Deepfake Vishing, Real-Time Voice Cloning Detection, LLM-Driven Impersonation},
        month = {September},
        }