Voice-Pro is an open-source, all-in-one multimedia processing web application designed for speech recognition, multilingual translation, and voice cloning. Developed by ABUS, the tool integrates a sophisticated AI stack, including Whisper, Faster-Whisper, and Whisper-Timestamped for high-accuracy automated speech recognition (ASR), alongside yt-dlp for YouTube video and audio extraction. For text-to-speech (TTS) and zero-shot voice cloning, the platform leverages F5-TTS, E2-TTS, CosyVoice (including the Fun-CosyVoice3-0.5B model supporting nine languages), and kokoro. Translation capabilities cover over 100 languages via Deep-Translator or optional Azure Translator integration. Recent version 4.0 updates transitioned the installer from Miniconda/pip to the uv package manager, significantly accelerating reproducible installations and upgrading the runtime to Python 3.12 and Torch 2.8.0+cu128. The application features a Gradio 6-based UI and is optimized for Windows with NVIDIA GPUs, supporting RTX 50-series hardware. It provides advanced features such as Demucs-based voice separation, subtitle file support (SRT, ASS, SSA), and customizable TTS parameters like pitch, volume, and speed. The project has been released as completely free and open-source software.