JefferyHcool/BiliNote

What it solves

BiliNote automates the process of taking notes from videos. It eliminates the need to manually transcribe and summarize long videos from platforms like Bilibili, YouTube, and Douyin by converting video content into structured, high-quality Markdown notes.

How it works

The tool extracts audio or subtitles from a provided video link and uses a transcription engine (such as Fast-Whisper or Groq) to convert speech to text. This text is then processed by a Large Language Model (like GPT, DeepSeek, or Qwen) to generate a structured summary. It can also automatically capture screenshots from the video to embed in the notes and uses RAG (Retrieval-Augmented Generation) to allow users to ask questions about the specific content of the generated notes.

Who it’s for

It is designed for students, researchers, and lifelong learners who consume educational or informational video content and want a searchable, organized text record of the key points without manual effort.

Highlights

• Multi-platform support: Works with Bilibili, YouTube, Douyin, Kuaishou, and local video files.
• Flexible AI Configuration: Supports various LLM providers (OpenAI, DeepSeek, Qwen) and multiple transcription engines (Fast-Whisper, MLX-Whisper, Groq).
• Interactive Notes: Includes automatic screenshots, links that jump back to the original video timestamp, and an AI Q&A system based on the note content.
• Deployment Options: Available as a cloud version, Docker container (with GPU acceleration support), and desktop clients for Windows and macOS.