Video Scribe is a web application that processes video files, extracting audio, transcribing speech, and capturing key frames. It provides a visual interface for analyzing video content and generating markdown-based transcripts with embedded images.
- Video upload and processing
- Audio extraction and transcription
- Key frame extraction
- Interactive transcript editing
- Markdown export with embedded images
- Dark mode support
- Node.js (v14 or later)
- FFmpeg
Clone the repository and install dependencies:
bash
git clone https://github.com/gramener/videoscribe.git
cd videoscribe
npm install- Start the server:
npm start- Open your browser and navigate to
http://localhost:3000. - Upload a video file and interact with the interface to process the video.
- Wait for the processing to complete. You'll see progress updates in the UI.
- Once processing is done, you can:
- View and edit the transcript
- Play the extracted audio
- Toggle key frames on/off
- Export the result as a markdown file
POST /audio: Extract audio from uploaded videoPOST /keyframes: Extract key frames from uploaded video
Both endpoints accept multipart form data with a file field containing the video file.
- Backend: Node.js, Express.js
- Frontend: HTML, CSS, JavaScript (ES6+)
- UI Framework: Bootstrap 5
- Templating: lit-html
- Audio Processing: FFmpeg
- Transcription: Groq API (distil-whisper-large-v3-en model)
This project is licensed under the MIT License.