We’re working on an exciting new addition to the SLR app: using AI to alter the spoken parts of our video content.
We plan to use AI tools to change the voice track of videos, by modulating the female voice so she can talk directly to you. We would also be able to add your voice so you can reply to her, making for an incredibly immersive real-life experience.
The task is to train AI so it can perfectly emulate a performer’s voice. If you’d like to know more about this process, please go here, where you can also hear some audio samples. This tool is EdiTTS, a speech editing program that can change content and pitch using AI technology.
You can see and hear it in action at this fascinating Youtube video where the performer sings and speaks in many different voices.
If you want to try for yourself, the Github code for real-time voice cloning is here.
This tool should work perfectly for our planned “build-your-own story” interactive videos, which you can read more about here.
We’d love to know your thoughts about this fascinating technology, and feel free to share your examples and ideas. Join in the discussion over at the SLR forum.