Whisper Audio/Video Transcription Tool 🎙️

A powerful Python-based transcription tool that leverages OpenAI's Whisper model to transcribe both audio and video files with GPU acceleration support.

🌟 Features

• Supports both audio and video file transcription
• Automatic audio extraction from video files
• GPU acceleration with CUDA support
• Timestamp-based transcription output
• Multi-language support
• Easy-to-use command line interface
• Saves transcriptions to text files

🔧 Usage Examples

Run the script with various arguments:

# Basic usage with a video file
python main.py --input video.mp4 --output transcript.txt

# Specify a different language (default is English)
python main.py --input audio.mp3 --language spanish

# Complete example with all options
python main.py --input interview.mp4 --output transcript.txt --language english

Available arguments:

• --input: Path to input audio/video file (required)
• --output: Path to output text file (optional, defaults to input filename + .txt)
• --language: Input language (optional, defaults to English)

🔧 Requirements

• Python 3.7+
• FFmpeg
• CUDA-compatible GPU (optional, for faster processing)
• Required Python packages (see requirements.txt)

🚀 Installation

1. Clone the repository:

git clone https://github.com/CarlosUlisesOchoa/Whisper-AI-write-video-or-audio-transcription-to-txt.git

2. Install the required Python packages:

pip install -r requirements.txt

3. Run the script:

py main.py --input "D:\files\myfile.mp4" --language en

You'll find the transcription in the same folder with the same name as the input file but with a .txt extension.

🔑 License

• GPL-3.0 license