Here is how you can clone your voice in seconds using OpenVoice. You can also convert text to speech using this application, and it is absolutely free.

First, we need to create a Python environment. In this video, we are using Anaconda. For those who do not have Anaconda, please visit anaconda.com and install it to your PC. After the installation is complete, you will find Anaconda Prompt in the Start menu. Create a Python environment by typing conda create –name openvoice python=3.10. After it has been installed, activate it with conda activate openvoice.

Now, go to the GitHub page and download all the OpenVoice files. Extract the ZIP folder to your desktop. In Anaconda Prompt, navigate to this folder using cd [folder path]. We will need to install all dependencies to run OpenVoice. For that, use pip install -r requirements.txt. Also, install FFmpeg using conda install ffmpeg. Since the application runs on Jupyter, install Jupyter Notebook with conda install -c conda-forge jupyterlab. You also need to install widgets using pip install notebook ipywidgets.

After this, download the V1 and V2 checkpoints from the OpenVoice GitHub page and extract them to the OpenVoice directory. To run the application, launch Jupyter Notebook with jupyter notebook, open demo_part1.ipynb, and execute each cell. Replace the sample audio in the resources folder with your own audio file, adjust parameters (e.g., style, speed), and generate cloned voice outputs in the output folder. For advanced features in demo_part3.ipynb, install MeloTTS via the provided GitHub command and resolve dependencies by running nltk_download.py.

To reopen OpenVoice after using it, open Anaconda Prompt, activate the environment, and navigate to the OpenVoice folder. After that, launch Jupyter Notebook. Ensure all files and checkpoints are present. You can then run demo1 or demo3 as per your requirements. We hope this video was useful.

Thank you very much.