No transcripts? No problem. HoverNotes watches the video directly and turns Bilibili lectures into structured notes — watch in any language and get your notes in the language you want.
Bilibili has some of the best educational content anywhere — programming, science, design, language learning — but taking notes on it is genuinely hard. Most videos have no transcript, much of the content is in Chinese, and the parts worth keeping (the code on screen, the diagram, the worked example) are exactly what a transcript would miss anyway.
What if the language and the missing transcript simply did not matter? You watch a Chinese programming tutorial and get clean, structured notes — code captured as timestamped screenshots, concepts organized, your own annotations added as you go. That is what becomes possible when the AI watches the video itself instead of relying on captions.
This guide walks through five ways to take notes on Bilibili, compares the tools honestly, and shows the AI-assisted approach that works without a transcript and across languages.
Bilibili is one of the largest repositories of educational video in the world, with deep tutorials, university lectures, and expert explanations — much of it unavailable anywhere else.
The catch: most of it has no transcript, and traditional note tools that depend on captions simply do not work here. You end up pausing constantly, trying to understand and type at the same time, and missing the visual content that carries most of the value.
The challenges
The challenge: Unlike YouTube, most Bilibili videos have no transcript, so transcript-based note tools are useless here.
The fix: HoverNotes watches the video directly, reading the on-screen content and listening to the audio to generate notes without needing a transcript.
The challenge: Much of the best content is in Mandarin, and even if you follow spoken Chinese you may want your notes in another language.
The fix: The AI watches the screen and hears the audio in any spoken language, then writes your notes in the language you choose — watch in Chinese, keep your notes in English.
The challenge: Bilibili tutorials lean hard on code, diagrams, and on-screen text — exactly what plain text notes miss.
The fix: Timestamped screenshot capture preserves the exact code and diagrams, linked back to the moment they were explained.
The challenge: Content is often organized into long series of 10–100+ videos, and notes need to connect across episodes.
The fix: Series-aware organization with cross-links between episodes builds one coherent knowledge base for the whole topic.
The methods
Pause often, use a translation tool for unfamiliar terms, and type notes by hand. The traditional route for language learners.
Pros
Cons
Best for: Chinese learners who want the language practice, on very short videos.
Screenshot everything important — code, diagrams, key slides — and add minimal annotations. Visual-first.
Pros
Cons
Best for: Programming tutorials where the value is on screen and you don't need the narration.
HoverNotes watches the video directly — no transcript required. It reads the on-screen content and the spoken audio, then writes structured, timestamped notes you can keep.
Pros
Cons
Best for: Almost all Bilibili content — especially when you don't speak Chinese or want clean notes fast.
Use Bilibili's bullet comments as crowd-sourced notes — viewers often flag key moments and add context.
Pros
Cons
Best for: Popular videos with active communities, as a supplement to real notes.
Turn key concepts into Anki flashcards for spaced-repetition learning, prioritizing recall over comprehensive notes.
Pros
Cons
Best for: Language learning, technical terms, and exam prep.
Tools compared
| Tool / approach | No transcript | Chinese video | Visual capture | Verdict |
|---|---|---|---|---|
| Manual typing | Yes | Slow | No | Too slow, misses visuals |
| Transcript-based AI | No | — | No | Doesn't work on Bilibili |
| Screenshot tools | Yes | — | Yes | No context, not searchable |
| Translation extensions | Partly | Manual | No | Word-by-word, no structure |
| HoverNotes | Yes | Yes | Yes | Built for it |
Why HoverNotes
Watches the video directly, so it works on Bilibili even though there's no transcript to lean on.
Watches and listens in any spoken language, then writes your notes in the language you choose — a Chinese lecture becomes clean English notes.
Timestamped screenshots of the code and diagrams that carry most of a tutorial's value.
Everything saves as Markdown to your own Obsidian vault on your machine — nothing stored in the cloud.
Build the system
Group notes by subject so material from multiple creators on the same topic lives together and stays searchable.
📁 Bilibili Learning/
📁 Programming/
- python-01-intro.md
- python-02-variables.md
📁 Language-Learning/
📁 Creative/
📁 _Topic-Index.mdConnect videos within a series and across topics with [[links]] and #tags, so related ideas are one click apart.
Screenshot code and diagrams with timestamps, and add your own translation of on-screen Chinese where it helps.
Revisit notes at the end of each series, and reference them while you build — notes you never review are just files.
Join thousands of students, professionals, and lifelong learners who use HoverNotes to enhance their video learning experience.
Questions
Use HoverNotes, which watches the video directly instead of relying on a transcript. The AI reads the on-screen content and the audio to generate structured notes, so it works on any Bilibili video whether or not a transcript exists.
Yes — in any language, both directions. The AI watches the screen and listens to the audio in any spoken language, then writes your notes in the language you want. Watch a Chinese-language Bilibili lecture and get clean, organized notes in English (or any language), with on-screen Chinese text, slides, and code captured along the way.
AI visual notes work best on Bilibili. Since most videos lack transcripts, an AI that watches the video directly is essential. HoverNotes captures timestamps, code, and on-screen visuals automatically while you add your own annotations.
Yes. HoverNotes captures code as timestamped screenshots exactly as it appears on screen, so it's especially useful for Bilibili's large library of programming tutorials.
Create folders by topic or series, with one note per video, and link related videos together. Since Bilibili content is often serialized, organizing by topic rather than uploader builds a more coherent knowledge base.
Locally and privately. HoverNotes saves your notes and timestamped screenshots as plain Markdown directly to your Obsidian vault on your computer. Nothing is stored in the cloud.
Keep going
The tool itself — take AI notes on any Bilibili video, including Chinese-language content.
Read moreThe complete guide to building a knowledge library from YouTube.
Read moreTurn any video — on any platform or a local file — into notes you keep.
Read moreNo transcripts, no Chinese fluency required. HoverNotes watches the video directly and turns it into notes you keep, saved to your Obsidian vault. Works on Bilibili, YouTube, Udemy, Coursera, and 20+ sites.
Free Starter includes 20 minutes of AI notes • Upgrade anytime