Creating a custom voice model on Kits.AI is simple with many different options. Professional Voice Cloning is our highest-quality cloning method requiring 15-30 minutes of an audio dataset and a few hours wait to process.
To create a Professional Voice Clone, head over to our train page and select what type of training you'd like to do. For the highest quality voice, select the Professional Voice Cloning option.
Select "Upload Files" (Don't have audio files? Choose the "Record from Scratch" option to enter our Guided Voice Cloning process)
You will now enter the file upload screen
Select the files you'd like to use to train your model. We recommend 10-60 minutes of dry (no effects) and monophonic (one note at a time) vocals.
Add your model details and continue
Confirm your model training and select "Train" to begin training your model
Depending on the size of your data, model training can take anywhere from 30 minutes to multiple hours. Follow along your AI voice generator's progress on your voices page
Read more on how you can create the best possible voice model here or watch our tutorial video: