Update README.md
Browse filesOpenVoice is an open-source initiative developed by a collaborative team of AI researchers from MIT, Tsinghua University, and the Canadian startup MyShell. The project’s primary goal is to create a versatile voice cloning tool that enables precise voice replication and natural-sounding speech generation across multiple languages and accents. Here are the key features of OpenVoice:
Accurate Tone Color Cloning:
OpenVoice excels at accurately cloning the tone color of a reference voice. It can faithfully reproduce the unique characteristics of a speaker’s voice, even when dealing with different languages and accents.
Flexible Voice Style Control:
Users have granular control over various voice styles, including emotion, accent, rhythm, pauses, and intonation. This flexibility allows for customized and expressive speech synthesis.
Zero-shot Cross-lingual Voice Cloning:
Unlike traditional methods, OpenVoice doesn’t require the language of the generated speech or the reference speech to be present in a massive multilingual training dataset. It can perform cross-lingual voice cloning effectively.
Cloning the Repository:
First, navigate to the /content directory using the %cd /content command.
Next, clone the GitHub repository named “OpenVoice” by running !git clone -b dev https://github.com/JoseRFJuniorLLMs/OpenVoice.
Change into the newly cloned repository with !cd /content/OpenVoice.
Installing Dependencies:
Install the aria2 package silently using !apt -y install -qq aria2.
Download the file checkpoints_1226.zip from the specified URL using !aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/JoseRFJunior/Giramille/resolve/main/checkpoints_1226.zip -d /content -o checkpoints_1226.zip.
Unzip the downloaded file with !unzip /content/checkpoints_1226.zip.
Python Libraries Installation:
Quietly install the following Python libraries:
gradio==3.50.2
langid
faster-whisper
whisper-timestamped
unidecode
eng-to-ipa
pypinyin
cn2an
Use !pip install -q followed by the library names.
Running the OpenVoice App:
Finally, execute the OpenVoice application with !python openvoice_app.py --share.