Video to Text Transcription Service

This project is a Flask-based web application that allows users to upload video files and receive transcriptions of the audio content. The application uses the Whisper model for transcription and MoviePy for audio extraction.

Acknowledgments

• Flask
• MoviePy
• Whisper

Features

• Video Upload: Users can upload video files through a web interface.
• Audio Extraction: The application extracts audio from the uploaded video.
• Transcription: The extracted audio is transcribed into text using the Whisper model.
• Download Transcripts: Users can download the transcribed text file.

Prerequisites

• Python 3.7+
• Flask
• Flask-HTTPAuth
• MoviePy
• Whisper
• Werkzeug

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/video-transcription-service.git
   cd video-transcription-service

2. Install dependencies:

   pip install -r requirements.txt

3. Set up user credentials:

   • Edit the app.py file to configure user credentials in the users dictionary.

4. Run the application:

   python app.py

5. Access the application:

   • Open your web browser and go to http://localhost:5000.

Usage

1. Upload a Video:

   • Navigate to the home page.
   • Select a video file to upload.
   • Click "Upload" to start the transcription process.

2. Download Transcript:

   • After processing, you will be redirected to a page where you can download the transcript.

To-Do

• Incorporate a way to reach the server over the internet, using ngrok.

File Structure

• app.py: Main application file containing the Flask app and routes.
• templates/upload.html: HTML template for the upload page.
• uploads/: Directory where uploaded videos are stored.
• transcripts/: Directory where transcribed text files are saved.

Security

• Basic HTTP authentication is implemented to secure the upload and download routes. Ensure to set strong credentials in the users dictionary.

License

This project is licensed under the MIT License. See the LICENSE file for more details.