An Effective Detection Of Voice Cloning Using Deep Learning

Voice cloning technology has been found applications in a wide ranges of areas varying from personalized speech interface to advertisements, video gaming, and something else. Exist voice cloning systems are capable of learn speaker characteristic from few samples and generated perceptually indistinguishable speech. This advance poses new security and privacies threat to voice-driven interfaces. The technology allows for the creation of synthesize, adjustable copied of a person's voice from recordings, which can be used to say anything typed into a systems. This has lead to its use in various legitimate fields, such as voiceover work and providing voices for those who can't speak. However, it also opens up possibilities for misusing in scam and frauds. This paper aim to identify such frauds of voice cloning and minimize the misusing.