Here’s how I cloned my voice using AI for free!

Published Jun 23, 2023

No headings found

Voice cloning software has existed since the '90s, and even before that. But while you had slim chances to see one work—or even better—clone your voice back then, such tools are far more accessible to the general public these days. And there are a ton of AI tools available online that can clone your voice within minutes (or hours). So let's take a look at one such free AI voice cloning tool.

What is AI voice cloning?

AI voice cloning or voice synthesis is a technology that uses machine learning / deep learning to recognize and replicate the unique patterns in an individual's voice. Over the past few years, there have been a lot of advances in this space leading to today where cloned voices can be extremely realistic with rich expressions and a natural tone. That being said, it is an unregulated space and we don't condone using this technology for the sake of spreading misinformation, spam, fraud, or hate speech.

Enter PlayHT, a free AI voice cloning tool

There are plenty of AI tools that let you create a clone of your voice. But since most of them are paid services, we decided to test something more accessible to everyone. And the software I used to clone my voice is called "PlayHT", which is a free AI voice cloning tool.

Here’s how you get started.

First, you go to their website "play.ht"
Then, sign up for an account
Now click on "Create Free AI voiceover"
- After that, you are greeted with the dashboard.
Select the "Voice Cloning" tab on your left and click "Create a New Clone"

Next, select the "Instant" option for free voice cloning.
- There's also "High Fidelity" cloning for better results, but that's only available to paid users.
Then upload a high-quality audio sample of the voice you'd like to duplicate.

And that's it! You can simply record yourself speaking for 30 – 120 seconds and upload that as the reference audio. After a few minutes, your voice is now ready to be used. Also Read:

My thoughts

Having also used Eleven Lab's AI voice cloning, I would say that it's in a league of its own when it comes to speech synthesis. And if you just want great text-to-speech results, then I would recommend you go there. But their voice cloning feature is behind a paywall so I guess it's not fair to compare it with this free AI voice cloning tool. And I think PlayHT's output is okay-ish and you can make it sound somewhat decent by cutting and pasting different parts. But it's definitely not perfect out of the gate. Maybe that's alright if you're just looking for some harmless fun or experiment. But for anything more than that, you have two options. Either: a) Pay for the "High Fidelity" cloning on PlayHT or subscribe to ElevenLabs' services or b) Generate tons of revisions and cut/paste them together for usable audio sample Here's a short sample of my cloned voice using this tool: [playlist ids="155661,155659,155658,155671"]