Self-host by2kb
v0.7.5Video-to-knowledge CLI for Bilibili/YouTube transcription
- Stars
- 156
- Requires
- Ffmpeg, SQLite
What is by2kb?
by2kb is a command-line tool for ingesting video content into a personal knowledge base. Given a Bilibili or YouTube URL (or local media file), it retrieves the transcript — preferring native captions, falling back to audio extraction and ASR via local faster-whisper or cloud Doubao — and produces three Markdown files: a verbatim timestamped transcript, a one-minute decision-oriented abstract, and a structured long-form study guide with knowledge map, claims, evidence, and action items. The tool is designed for agent-first workflows: a Hermes plugin intercepts video URLs shared to Telegram, runs ingestion in the background, and returns the artifacts via the same chat. Users without an agent can run the same pipeline with an OpenAI-compatible API key. All artifacts are stored as portable Markdown and JSON in a local directory (default ~/.by2kb) or an Obsidian vault. The project uses SQLite for job tracking and deduplication, requires Python 3.12+, pipx, ffmpeg, and optionally Deno/Node for YouTube player challenges. No resident service or Docker image is provided; it runs as a local CLI subprocess.
Should you self-host by2kb?
by2kb scores 43/100 on selfhost.directory (how scoring works) and ranks #27 of 42 in Knowledge Management. It has 156 GitHub stars.
It shipped 10 stable releases in the past year. The latest, v0.7.5, came out on 19 October 2026. New versions land roughly every 2 days.
Setup is moderate: expect some configuration beyond a single command. It is MIT-licensed and written mostly in Python.
Its closest alternative on selfhost.directory is Project N.O.M.A.D. (85/100, 41 075 stars). It scores higher overall, so compare the two before you pick. See by2kb vs Project N.O.M.A.D. side by side.
Features
- Agent-first Hermes integration
- Cloud Doubao ASR
- Content-addressed deduplication
- Local faster-whisper ASR
- Local media ingestion
- Obsidian vault output
- Personalized skills
- Short abstract artifact
- Standalone CLI mode
- Video transcription
by2kb README
More in Video Processing
CliparrSelf-hosted, open-source tool that clips video from Plex, Jellyfin, or local files directly in your browser — no server-side FFmpeg needed.
ClipForgeSelf-hosted AI turns product images into short videos for Douyin, Kuaishou, Xiaohongshu, TikTok Shop — product-preserving visuals, AI scripts, voiceovers, subtitles.
cobaltA self-hosted, open-source media downloader that saves video, audio, gifs and photos from YouTube, TikTok, Instagram, Reddit and more, with no ads or trackers.
ConcatAn open-source, self-hosted desktop and mobile video editor pitched as a CapCut replacement, with local Whisper captions, text-to-speech and 4K export.
DigicutOpen-source, self-hosted web app that cuts an excerpt out of an audio or video file in the browser with FFmpeg.wasm, so media never reaches a server.
DigimergeSelf-hosted open-source (AGPL-3.0) web app to merge audio, concatenate videos or build image slideshows with ffmpeg.wasm; files are processed locally in the browser.
DigitranscodeSelf-hosted audio and video converter that transcodes in the browser with FFmpeg.wasm — nothing is uploaded. Offers bitrate, sample-rate and resolution options.
dovi-convert-dockerA self-hosted Docker app with a browser interface that turns Dolby Vision Profile 7 rips into Profile 8.1 so they play on more devices.
Duix.AvatarSelf-hosted, open-source AI avatar toolkit that clones your face and voice offline to generate lip-synced talking videos from text or audio on your own GPU.
EchoSyncEchoSync is a self-hosted AI video generator turning text into multi-scene prototypes with narration, subtitles, and a digital anchor — all running locally via Docker.
ffmpeg-over-ipOpen-source, self-hosted client and server that run GPU-accelerated ffmpeg on a remote machine and tunnel file I/O back, dropping GPU passthrough and NFS/SMB shares.
FlyReq Image StudioFlyReq Image Studio is a self-hosted, open-source AI image and video generation workspace with multi-model support, infinite canvas, agent mode, and PWA.
0 comments