Deep Learning-Based Analysis of a Real-Time Voice Cloning System

Voice technology has emerged as a hotspot for<br> deep learning research due to fast advancements in<br> computer technology. The goal of human-computer<br> interaction should be to give computers the ability to feel,<br> see, hear, and speak. Voice is the most favorable<br> approach for future interactions between humans and<br> computers because it offers more benefits than any other<br> method. One example of voice technology that is capable<br> of imitating a particular person's voice is voice cloning.<br> Real-time voice cloning with only a few samples is<br> proposed as a solution to the issue of having to provide a<br> large amount of samples and having to endure a long time<br> in the past for voice cloning. This strategy deviates from<br> the conventional model.<br> For independent training, different databases and<br> models are used but for joint modeling, only three models<br> are used. The vocoder makes use of a novel type of LPCNET that works well on certain samples and low-<br> performance devices.