JefferyHcool/BiliNote
BiliNote
BiliNote is a tool that converts video URLs into structured, organized notes. It works by extracting video content and metadata from major platforms, transcribing audio to text entirely on-device using a local speech recognition model, and then summarizing the transcript with a language model to produce clean notes that can include screenshots and timestamp links.
What sets BiliNote apart is its configurable AI backend, which lets you choose and switch between different language model providers for generating summaries. All transcription happens offline and locally, preserving privacy and enabling use without an internet connection. The tool also maintains a versioned history of every note generated, allowing you to review, compare, or restore earlier versions. For further exploration, generated notes are indexed into vector embeddings, enabling you to ask natural language questions and retrieve relevant passages with answers via RAG.
Beyond its core workflows, BiliNote supports content capture through a browser extension and provides a unified pipeline for ingesting video from multiple platforms, extracting metadata, and coordinating transcription and summarization into a single automated process.
Features
- Summarizations - Produces organized notes from video links by extracting content, transcribing, and summarizing with a language model.
- Local Speech-to-Text - Transcribes audio to text entirely on-device using a local speech recognition model without cloud dependency.
- Multi-Platform Video Note Generators - Extracts content from videos on major platforms and produces organized notes automatically.
- Video Note Generators - Automates the complete video-to-note workflow: extracts content, transcribes, and generates structured notes.
- Structured Video Note Generators - Accepts video links from any platform and outputs structured notes with screenshots and timestamp links.
- Video Note Generators - Extracts content from video URLs, transcribes audio, and summarizes with a language model to produce structured notes.
- Video-to-Note Pipelines - Orchestrates subtitle download, on-device transcription, metadata capture, and LLM summarization into a single sequential workflow for producing structured notes from videos.
- Multi-Platform Video Extraction - Enables parsing video links from major platforms and standardizing content extraction from diverse sources.
- AI Backend Integrations - Allows users to select and configure different AI services for content understanding and note generation.
- AI Service Configurations - Allows selecting and configuring different AI service providers for content summarization and note generation.
- Embedding-Based Retrieval - Indexes generated notes into vector embeddings to enable semantic search and retrieval of relevant passages.
- Question Answering Systems - Indexes generated notes to allow asking natural language questions and retrieving relevant passages with answer generation.
- Embedding-Based Note Searches - Indexes generated notes into vector embeddings for answering natural language questions with retrieved passages.
- Natural Language Search Interfaces - Enables querying note content using natural language with passage retrieval and answer generation.
- Semantic Note Retrieval - Indexes generated notes to answer natural language questions by retrieving relevant passages and generating responses.
Star history
How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.
