A Streamlit-based web application that extracts and summarizes the transcript of a YouTube video using Google Generative AI (Gemini) API.
This tool helps users quickly understand the contents of a YouTube video without watching the entire content. It is ideal for researchers, students, content creators, and busy professionals who want quick insights from educational or lengthy videos.
- User pastes a YouTube video link into the app.
- The transcript is automatically fetched using the YouTube Transcript API.
- The transcript is then passed to Google's Gemini (Generative AI) API.
- Gemini generates a concise and informative summary (within 250 words).
- The summary is displayed on the page alongside the video thumbnail.
- Create environment: conda create -n yt-summarizer python=3.10 conda activate yt-summarizer
- Install dependencies: pip install -r requirements.txt
Create a "requirements.txt" file with the following content: streamlit python-dotenv google-generativeai youtube-transcript-api
-
Create a
.env
file in the project root: touch .env -
Add your Google API key: GOOGLE_API_KEY=your_google_generative_ai_api_key_here
streamlit run app.py
- Summarizes YouTube video transcripts into bullet points or short paragraphs.
- Displays the video thumbnail.
- Uses Google's Gemini (Generative AI) for intelligent summarization.
- Lightweight and easy-to-use web interface with Streamlit.
- Saves time by summarizing long videos.
- Requires no video watching—only the transcript.
- Efficient for content review, note-taking, and research.
- Easy to deploy on platforms like Streamlit Cloud or Hugging Face Spaces
- Allow user to select summary length or format (e.g., bullets, paragraph, outline).
- Add multilingual transcript support.
- Option to download summary as PDF or TXT.
- Archive and manage past summaries.
- Display full transcript alongside summary.