I just cloned my voice with Generative AI. Please take a listen to my AI-generated reading of Lewis Carroll and let me know what you think!
Background: I’ve been wanting to start a podcast for 20 years but could never seem to get to it. The need to have quiet surroundings, proper audio equipment, and a lot of patience in the editing process always made my spirit falter.
So, I decided to try out ElevenLabs’ professional voice cloning with their paid subscription tier ($22 monthly). They recommended uploading 3 hours of my recorded audio, but I found it exhausting just recording 1.5 hours of me reading a Jim Rohn book aloud.
It took about half a day for ElevenLabs to process and finetune my AI voice, and voilà, I can now produce a clean audio recording that doesn’t have my usual “err” and “you know?” filler words and does not need any editing. I do think I need to finetune my voice further by uploading more voice samples.
Of course, this will raise your alarms immediately about deepfakes! Let’s discuss that in the comments.