TikTok is developing a new AI-based feature to clone users' voice

The latest version of the TikTok app for Android contains lines detailing the brand new artificial intelligence-based text-to-speech feature. This technology gives users the ability to create unique voice recordings that sound like their personal voice.

Here's What We Know

The essence of the technology is that the user records a sample of their voice and then the AI analyses and plays it back, creating realistic voice messages. This opens up new possibilities for content creation, as users can generate voice recordings without having to speak in front of a microphone themselves.

Users who want to use this feature will have to read and accept the terms of use of the "TikTok Voice Library". They will then be able to record their voice, which will be processed and converted into an artificial voice, which will take 10 seconds to complete. This voice can then be used in the text-to-speech feature to create videos in TikTok.

In terms of privacy, users can rest assured that their artificial voice will remain private and can be deleted at any time. Additionally, users will be able to remove their artificial voice from the "Voice Library," which will prevent other users from using it in their content.

Unfortunately, the feature is not fully ready at the moment and some of its elements, such as the text to read when recording a voice, are not yet available. However, when it goes live, users will be able to evaluate how well it handles voice cloning compared to other similar technologies.

Source: TheSpAndroid