1. Introduction to BiliNote
What is BiliNote?
BiliNote is an open-source AI video note-taking assistant designed to simplify content extraction from videos on platforms like Bilibili and YouTube. By converting video content into structured, Markdown formatted notes, BiliNote enhances efficiency for users looking to organize and review video materials effortlessly. With features such as insertable screenshots, jump links to original videos, task logging, and historical review, BiliNote serves as a comprehensive tool for managing video-based information.
Why Do We Need BiliNote?
In today’s information-rich world, efficiently extracting core insights from voluminous video content poses a significant challenge. Traditional note-taking methods are time-consuming and labor-intensive. BiliNote addresses this issue by leveraging AI technology to automate content processing, thereby improving both the speed and quality of note generation.
How to Get Started with BiliNote?
To experience BiliNote firsthand, visit the official website: https://www.bilinote.app. Note that since the service is hosted on Cloudflare Pages, there might be occasional delays in loading times.
2. Key Features of BiliNote
Multi-Platform Support
BiliNote currently supports Bilibili and YouTube, with plans to expand to more platforms in the future. This means无论您使用哪个平台观看视频，都可以通过BiliNote进行笔记生成。
Local Model Audio Transcription
BiliNote utilizes Fast-Whisper for audio transcription, ensuring accurate conversion of spoken content within videos.
GPT Model for Video Summary
Leveraging advanced models from OpenAI, DeepSeek, and Qwen, BiliNote can summarize video content intelligently, producing high-quality notes.
Structured Markdown Notes
BiliNote converts video content into well-structured Markdown formatted notes, facilitating easy review and editing.
Optional Screenshot Integration
Users can optionally insert video screenshots into their notes, making them more visually engaging.
Jump Links for Content Navigation
BiliNote supports the creation of jump links, enabling direct access to corresponding parts of the original video.
Task Logging and Historical Review
Automatic task logging and historical review capabilities allow users to keep track of their note-taking activities and review past notes anytime.
3. Getting Started Quickly
Clone the Repository
git clone https://github.com/JefferyHcool/BiliNote.git
cd BiliNote
mv .env.example .env
Start the Backend (FastAPI)
Navigate to the backend directory and install dependencies:
cd backend
pip install -r requirements.txt
uvicorn app.main:app --reload
Start the Frontend (Vite + React)
cd BiliNote_frontend
pnpm install
pnpm dev
Access the interface at: http://localhost:5173.
4. Dependency Requirements
FFmpeg
FFmpeg is essential for audio processing and transcoding in BiliNote. Please install it according to your operating system:
-
Mac (brew): brew install ffmpeg -
Ubuntu / Debian: sudo apt install ffmpeg -
Windows: Download and install from the official website: https://ffmpeg.org/download.html
CUDA Acceleration (Optional)
For faster audio transcription tasks, use a machine equipped with a NVIDIA GPU and enable the fast-whisper + CUDA accelerated version. For configuration details, refer to: fast-whisper project page.
Docker One-Click Deployment
Ensure Docker and Docker Compose are installed:
1. Clone this project
git clone https://github.com/JefferyHcool/BiliNote.git
cd BiliNote
mv .env.example .env
2. Start Docker Compose
docker compose up --build
Default ports:
-
Frontend: http://localhost:${FRONTEND_PORT} -
Backend: http://localhost:${BACKEND_PORT}
Ports and environment settings can be customized in the .env file.
5. Environment Variables Configuration
Here is a sample configuration for the backend .env file:
API_BASE_URL=http://localhost:8000
OUT_DIR=note_results
IMAGE_BASE_URL=/static/screenshots
MODEl_PROVIDER=openai
OPENAI_API_KEY=sk-xxxxxx
DEEP_SEEK_API_KEY=xxx
QWEN_API_KEY=xxx
6. Future Development Plans
Expanding Supported Platforms
BiliNote aims to support more video platforms like Douyin and Kuaishou, catering to a broader user base.
Frontend Model Switching
Users will be able to switch between different AI models and speech-to-text models directly through the frontend settings.
Customizable AI Summary Styles
Multiple AI summary styles—academic, conversational, key point extraction—will be available, accommodating various note-taking scenarios.
Export Notes to PDF/Word/Notion
The ability to export notes to PDF, Word, and Notion formats will be added, enhancing versatility across different platforms.
Additional Model Support
More advanced AI models will be incorporated to improve note generation effectiveness and efficiency.
Expanded Audio-to-Text Model Support
Enhanced support for additional audio-to-text models will be introduced to boost speech recognition accuracy and diversity.
7. Community Engagement
BiliNote Q&A QQ Group
Join the BiliNote Q&A QQ group at 785367111 to engage with other users and share experiences.
BiliNote WeChat Group
Details about the BiliNote WeChat group can be found on the project page.
8. License Information
BiliNote is licensed under the MIT License, allowing free usage, modification, and distribution of the code. Your support and feedback are driving forces for our continuous improvement! We welcome pull requests, issues, and stars!
