Whisper Transcription Tool

A simple GUI application that uses OpenAI's Whisper API to transcribe audio files into text.

Features

• Transcribe audio files in formats like MP3, WAV, M4A, and FLAC.
• Automatically compresses files larger than 25 MB.
• Displays transcription results within the application.
• Copy transcription text to the clipboard.

GUI

Requirements

• Python 3.11.10
• tk
• requests
• ffmpeg (for audio compression)
• yt-dlp (for video downloading)

Installation

1. Clone the repository:

   git clone github.com/awerks/whisper_gui.git

2. Navigate to the project directory:

   cd whisper_gui

3. Create and activate the Conda environment:

   conda env create -n transcription -f environment.yml
   conda activate transcription

4. Install ffmpeg:

   • On macOS using Homebrew:

     brew install ffmpeg

   • On Windows, download from FFmpeg website.

5. Install yt-dlp:

   • On macOS using Homebrew:

     brew install yt-dlp

   • On Windows, download from yt-dlp releases.

6. Set up the OpenAI API key:

   • Obtain your OpenAI API key and set it as an environment variable:

     • On Linux/macOS:

       export OPENAI_TRANSCRIPTION_KEY='your_api_key_here'

     • On Windows:

       set OPENAI_TRANSCRIPTION_KEY='your_api_key_here'

Usage

Run the application:

python3 program.py

• A GUI window will open where you can select an audio file and transcribe it.
• Transcription results will appear in the text area of the application.
• Use the "Copy to Clipboard" button to copy the transcription.

Notes

• Ensure ffmpeg is installed and added to your system's PATH.
• The application supports audio files up to 25 MB without compression.
• Larger files will be compressed automatically before transcription.

License

Distributed under the MIT License.

Acknowledgments

• Powered by OpenAI's Whisper API.
• FFmpeg.
• yt-dlp