Media to Text Converter

A Flask web application that converts audio and video files to text using OpenAI's Whisper model.

Features

• Upload multiple media files at once
• Drag and drop file upload interface
• Real-time processing progress tracking
• Support for various audio/video formats (MP3, MP4, WAV, M4A, etc.)
• Download transcribed text as .txt files
• Copy text to clipboard functionality
• Responsive web interface

Installation

1. Install Python dependencies:

pip install -r requirements.txt

2. Install FFmpeg (required by Whisper):

# Ubuntu/Debian
sudo apt update && sudo apt install ffmpeg

# CentOS/RHEL
sudo yum install ffmpeg

# macOS
brew install ffmpeg

Usage

1. Run the Flask application:

python app.py

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

3. Upload one or more media files using the drag-and-drop interface

4. Wait for the conversion to complete and view/download your transcribed text

Supported File Formats

• Audio: MP3, M4A, WAV, MPGA
• Video: MP4, MPEG, WebM, MOV, AVI, FLV, MKV

Configuration

You can modify the Whisper model in app.py by changing the model name in the load_whisper_model() function:

• tiny.en - Fastest, English only
• base.en - Better accuracy, English only
• small.en - Good balance of speed and accuracy
• medium.en - Higher accuracy
• large - Best accuracy, supports multiple languages

API Endpoints

• GET / - Main upload page
• POST /upload - Upload files for processing
• GET /status/<job_id> - Check processing status
• GET /result/<job_id> - View transcription result
• GET /download/<job_id> - Download transcription as text file

Notes

• Maximum file size: 500MB
• Files are processed asynchronously in the background
• Uploaded files are automatically deleted after processing
• Transcribed text files are saved temporarily for download