Flask Video Summarizer — From Video to Transcript
W A Flask-based web application that lets users upload a video file, automatically extracts the audio, transcribes speech into text using Facebook’s Wav2Vec2 ASR model, and summarizes it with the BART text summarization model. The app displays both the complete transcript and a concise summary in the browser, providing an end-to-end video-to-text processing solution.
Project Summery
A simple Flask application that allows users to upload a video file and:
- 🎧 Extracts audio from the video using
moviepy - 🧠 Transcribes speech into text using
facebook/wav2vec2-base-960h(ASR model) - ✍️ Summarizes the transcript using
facebook/bart-large-cnn(text summarization model)
Both the full transcript and the summary are displayed in the browser after processing.
🚀 Features
- ✅ Upload video files (
.mp4,.mov,.avi, etc.) - 🎧 Automatically extract audio from video
- 🔄 Resample audio to 16kHz with
librosa - 🧠 Transcribe using Hugging Face's Wav2Vec2 model
- ✍️ Summarize the transcript with BART
- 🌐 Simple web interface using Flask
🛠 Setup Instructions
1. Clone the Repository
git clone https://github.com/sujit-codezen/video_summary.git cd video_summary
2. Create Python Environment & Install Dependencies
python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate pip install -r requirements.txt
3. Install Requirements
pip install -r requirements.txt
4. Running the App
python app.py
🧠 Models Used
| Task | Model | Source |
|---|---|---|
| Transcription | facebook/wav2vec2-base-960h |
Hugging Face 🤗 |
| Summarization | facebook/bart-large-cnn |
Hugging Face 🤗 |
Output
