Clone your voice in 5 minutes!

Pawel Stasinski
Oct 29 · 2 min read

I want to show you an excellent library to clone your voice. I mean that machine could read a text using your voice!

Let's start!

In this article, which I am using code from the repository :

and his forked version(for CPU):

Its video will show you what it’s all about.

Steps to run minimal example(with supported CPU)

#terminal
#you could find requirements.txt in this repo
https://github.com/fuwiak/medium/tree/master/5minutes
virtualenv work
source work/bin/activate
pip3 install -r requirements.txt

http://www.openslr.org/resources/12/train-clean-100.tar.gz

After unpacking you should see a directory with name LibriSpeech

Assuming that directory LibriSpeech is in Desktop:

#terminalgit clone https://github.com/shawwn/Real-Time-Voice-Cloning
cd Real-Time-Voice-Cloning
python demo_toolbox.py -d ~/Desktop/LibriSpeech/train-clean-100/

After its steps, you should see something like this:

After recording own voice(button Record) you should see something like this:

To read the text “Welcome to the toolbox …” push Synthesize and vocode.


That's all folks! In any case, I look forward to feedback.

Pawel Stasinski

Written by

Freelancer Machine Learning/Data Science at Upwork https://www.upwork.com/fl/pawelstasinski2 https://www.linkedin.com/in/pawe%C5%82-stasinski-7a429311a/

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade