Video Transcription and Subtitling using ML

This project involves converting video speech to text using the Whisper ASR model and generating subtitles. The code is designed for a specific use case and includes features such as splitting the text into lines, creating captions, and overlaying them on the original video.

Table of Contents

• Setup
• Usage
• Dependencies
• License

Setup

1. Clone the repository: git clone <repository-url>
2. Install the required dependencies using the provided pip installations and apt-get commands.
3. Mount Google Drive to access video files: drive.mount('/content/drive')

Usage

1. Provide the path to the input video: input_video_path = "/content/drive/MyDrive/video/sample.mp4"
2. Run the notebook to convert the video to text, split it into lines, and generate subtitles.
3. Adjust parameters such as MaxChars, MaxDuration, and MaxGap in the code as needed.
4. The final output video with subtitles will be saved as output.mp4 in the specified Google Drive location.

Dependencies

• faster-whisper
• moviepy
• imageio

Make sure to install these dependencies before running the code.