Voice-Pro: open source multifunctional video translation tool, voice transcription and translation into multiple languages, Windows one-click installation

General Introduction

Voice-Pro is a multifunctional tool based on Gradio WebUI that supports speech-to-text, text-to-speech, real-time translation, YouTube video downloads, and human voice separation. It integrates Whisper, Faster-Whisper and Whisper-Timestamped technologies to provide efficient audio processing and translation for multiple languages and scenarios.

Function List

• speech-to-text: Supports Whisper, Faster-Whisper, and Whisper-Timestamped to provide highly accurate speech recognition.
• text-to-speech: Supports Edge-TTS and F5-TTS with multiple language and voice choices, speed, volume and pitch adjustments.
• real time translation: Supports real-time speech recognition and translation for multiple languages.
• YouTube Download: You can download YouTube videos and extract audio (mp3, wav, flac).
• vocal separation: Vocal and background sound separation using MDX-Net and Demucs engines.
• batch file: Supports subtitle generation, translation and text-to-speech processing of large batches of files.
• Subtitle Generation: Supports generation and editing of subtitles in more than 90 languages.
• Multi-format support: All video and audio formats supported by ffmpeg are supported.

Using Help

Installation process

1. starter pack: Clone or download the latest version of the source code from GitHub.

   git clone https://github.com/abus-aikorea/voice-pro.git

2. Install and run the program::
   • (of a computer) run configure.bat Install the required dependencies (e.g. git, ffmpeg and CUDA).
   • (of a computer) run start.bat Start Voice-Pro and WebUI will run automatically.
   • When run for the first time, Voice-Pro will first install, which may take an hour or more, during which time do not close the Windows command window.

Usage Functions

1. speech-to-text::
   • On the Studio tab, select Whisper Models and types of calculations.
   • Upload an audio file or select an audio input source (such as a microphone).
   • Click the "Start" button and wait for the speech recognition and subtitle creation to complete.
2. rendering::
   • Upload the text or subtitle file to be translated in the Translate tab.
   • Select the target language and click the "Translate" button.
   • Once the translation is complete, you can download the translated file.
3. text-to-speech::
   • Select Edge-TTS or F5-TTS in the TTS tab.
   • Enter the text to be converted and select the speech parameters (e.g. speed, volume, pitch).
   • Click the "Generate Voice" button and wait for the voice generation to complete.
4. YouTube Download::
   • Enter the YouTube video link in the YouTube Downloader tab.
   • Select the audio format (mp3, wav, flac) and click the "Download" button.
   • Once the download is complete, you can find the audio file in the specified folder.
5. sound separation::
   • Upload audio files in the Vocal Remover tab.
   • Select the MDX-Net or Demucs engine and click on the Start button.
   • Wait for the sound separation to complete and you can download the separated audio file.
6. batch file::
   • Upload multiple files in the Batch tab.
   • Select the desired operation (subtitling, translation, text-to-speech).
   • Click the "Start" button and wait for batch processing to complete.

common problems

• Browser not running automatically: Close the Windows command window and re-run start.bat, or manually enter the displayed address in your browser (e.g. http://127.0.0.1:7892).
• CUDA Out of Memory Error: Check the GPU memory status to adjust the noise reduction level or calculation type.
• Windows Defender Warning: Add the batch file as an exception or temporarily disable Windows Defender.