Clone Any Voice In 15 Seconds Using E2 F5 TTS

In this short video I show how to setup Windows and Audacity to record the sound coming out of your speakers using Stereo Mix. Then a short demo of E2-F5-TTS cloning Jim Al-Khalili's voice and recite a short speech.

Apps used in this video

Pinokio

Once Pinokio is installed, click Discover and then install E2-F5-TTS. It may take a few minutes to install the first time.

Audacity

Record your own voice or use your own voice.

I also use a free Plug-in for Audacity by Intel called OpenVino.
This plug in adds two options to the Effects Window
1. Music Separation
2. Noise Suppression

OpenVino for Audacity

Run the E2-F5-TTS app and upload your source audio.

Type in your text to generate and click 'Synthesize'.

Optional, but I have found it helps, set the speed to .9.

Depending on how your output audio is, this may take a minute or two or longer.

I found E2-F5-TTS to be super useful, let me know what you think.

[AnimagicToons]

Very clear tutorial. Thanks! The tips on how to record audio that plays on the PC were very helpful.

[ResellDecadence]

Great insight, I am unable to do it …

[BVLVI]

I can't find samples to use anywhere. I just want some sample voices to play around with it. Can't believe there isn't tts sound packs.. if anyone has some please shoot me a link please and thank you.

[popeopera]

I get this when attempting to run: {ENOENT: no such file or directory, stat
And the files are in that location...🤔

[mercuryin]

My question is if I want to use the cloned voice as tts with for example home assistant how I can export the models to be used without the program, I mean to be able to tts with that cloned voice all the time ? Make sense ?

[frailalbatross]

can the sample using another language? My synthesized voice don't have a sound