Key Features
- Voice Changer (RVC): Transform your voice with real-time conversion.
- Zero-shot Voice Cloning (E2, F5-TTS): Clone any voice with just a short sample.
- YouTube Downloading: Extract audio from videos in multiple formats (mp3, wav, flac).
- Vocal Isolation (UVR5): Separate vocals from background using MDX-Net and Demucs.
- Text-to-Speech (Edge-TTS): Generate natural-sounding speech in multiple languages.
- Multi-language Translation: Instantly translate audio and subtitles into various languages.
Who Can Benefit?
- Content Creators: Create dubbed videos or podcasts with cloned voices.
- Researchers: Analyze and process audio data with advanced transcription tools.
- Multilingual Professionals: Translate and subtitle media in real-time.
Technical Highlights
- Powered by advanced Whisper engines: Whisper, Faster-Whisper, and Whisper-Timestamped for top-tier speech recognition.
- Supports NVIDIA GPUs with CUDA 12.1 for enhanced performance.
- Minimum 4GB VRAM required; 8GB recommended for optimal processing.
Celebrity Voice Samples
Explore how Voice-Pro can clone celebrity voices with zero-shot technology.
Installation
- Clone or download from GitHub:
git clone https://github.com/abus-aikorea/voice-pro.git. - Run
configure.batto set up the environment. - Launch the WebUI by running
start.bat. - Requires Windows 10/11 (64-bit), NVIDIA GPU with CUDA 12.1, and 4GB+ VRAM (8GB recommended).
Explore Voice-Pro
Demo Videos
Click the thumbnails to watch Voice-Pro in action!
Links
Contact
Email: abus.aikorea@gmail.com
Voice-Pro in the Community
Voice-Pro has been recognized and discussed across various platforms. Here are some notable mentions:
- Hacker News: Featured in "Show HN: Voice-Pro – AI Voice Cloning" for its open-source audio manipulation capabilities.
- SourceForge: Hosted as a mirror with user reviews, highlighting its transcription and translation features.
- AIShareNet: Introduced as a versatile audio processing tool with real-time translation and YouTube downloads.