How I made voice cloning accessible to everyone on Discord

Hey Indie Hackers,

I've been fascinated by voice cloning for quite some time now. It has advanced at a rapid pace over the past several years, but there was one glaring issue with the industry around it... The ease of use for the average person.

If someone wanted to try out voice cloning, they would have to visit some website they've most likely never heard of, figure out how to record a clip of their voice, upload that clip, and finally download an audio file containing the cloned speech. That's a lot of friction for most people.

Several months ago, I thought to myself, "What if voice cloning could be directly integrated into a social platform like Discord?" Excited by the thought, I immediately began working on a prototype for what is now known as CopyKitten.

CopyKitten is a Discord bot that will join your voice call, listen to you speak, and create a clone of your voice. No external website, all integrated directly within Discord. I've tested just about all of the AI voice cloning model providers, and for this project, I decided to go with FishAudio for its great balance of quality, realism, speed, and cost.

I recently released CopyKitten, and things have been going great so far. Plenty of user feedback has made it my way, and I'm always working to further improve the core user experience. If this interests you at all, I would love it if you tested it for yourself and let me know what you think!