Video Subtitle Extractor Bot

A simple Telegram bot to extract hard-coded subtitle from videos using FFmpeg & Tesseract.

Note that the recognition accuracy depends on the subtitle font and its visibility.

Deploy to Heroku

Local Deploying

1. Make sure you have Tesseract and FFmpeg.

2. Clone the repo

   git clone https://github.com/samadii/VidSubExtract-Bot
   

3. Go to this line and add the PATH where Tesseract is installed.

4. Also fill this path with the path of the tessdata folder.

5. Enter the directory

   cd VidSubExtract-Bot
   

6. Install all requirements using pip.

   pip3 install -r requirements.txt
   

7. Run the file

   python3 main.py
   

Environment Variables

• API_ID - Get this from my.telegram.org
• API_HASH - Get this from my.telegram.org
• BOT_TOKEN - Get this from @BotFather
• SUBTITLE_LANG - Get this from list of ISO 639-2 language codes
• USE_CROP - Set this to ANYTHING to enable crop mode(crop out subtitles area).

Devs:

• @samadii