The AI workspace for your YouTube videos

YouTube is the largest library of knowledge on the internet, and almost none of it is searchable once you've watched it. A two-hour conference talk contains a ten-minute section that's exactly what you need for a project. A tutorial series walks through a technique across six videos. A researcher gives an interview that reframes how you think about a problem. A founder explains their growth strategy in a podcast episode posted as a video. You watch it. You learn something. And then it's gone, buried in your watch history or a playlist you'll never revisit, because the only way to find a specific moment is to press play and scrub until you recognise it.

Fabric transcribes YouTube videos and makes the spoken content searchable by meaning. Find the exact moment from a two-hour talk in seconds. Ask the AI questions about what was said. Embed videos on the canvas alongside your notes and research. Build a searchable library from the best of YouTube and connect it to everything else you know.

Transcribe and search any YouTube video

Save a YouTube video to Fabric and audio and video transcription generates a timestamped, searchable transcript. The speaker's argument, the tutorial's steps, the interview's key insight, all turned into text you can search, highlight, and reference.

AI search reads every transcript and searches by meaning. Ask "the part where she explains transformer attention mechanisms" or "when the founder discussed pricing their first product" and find the exact moment with a timestamp. No scrubbing. No re-watching. The search works by concept, so it finds the right section even when the speaker used different phrasing from your query.

Search works across every video you've saved and across every other content type in your workspace. A concept from a YouTube talk is findable alongside the article that covers the same topic and the notes you wrote about it.

Ask the AI about what you've watched

The AI assistant treats your saved YouTube transcripts as a knowledge base. Ask it to summarise a two-hour conference talk in three paragraphs. Ask it to extract the five key takeaways from a tutorial series. Ask it to compare what two speakers said about the same topic across different talks. Ask it to find every video you've saved where someone discussed a specific technique.

For heavy YouTube learners, this changes the medium entirely. Instead of passive consumption that fades, the content becomes an active, queryable part of your knowledge. "What have the founders I've watched say about product-market fit" draws from across dozens of videos and produces a synthesis that no single watch could provide.

The assistant works from video transcripts alongside your documents, notes, articles, and other materials. It can connect what was said in a YouTube video to what you've read in an article or written in your own notes, which is how learning actually works: across formats, not within them.

Embed YouTube on the canvas

The canvas supports live YouTube embeds. Videos are playable directly on the canvas alongside images, notes, articles, and other references. Build a reference board for a project with key talks embedded alongside written sources. Plan a content calendar with inspiration videos next to your drafts. Create a study board with lecture videos, reading notes, and your own thinking arranged spatially.

This is especially useful for moodboards and brainstorming, where a video reference sitting next to a still image and a note tells a story that a playlist can't.

Notes alongside the video

Watch a YouTube video and write your thoughts in notes and docs alongside the transcript. Your own observations, the key points you took away, the connections you drew to other things you know, all searchable alongside the video content.

Annotations let you highlight and comment on the transcript itself. Mark the key argument. Flag the step in the tutorial. Note where the speaker contradicts something from another video. Your annotations are searchable across your full library, so a note on a transcript from six months ago is findable when the concept becomes relevant again.

Subscribe to channels in your workspace

Follow YouTube channels and creators with RSS feeds. New uploads arrive in your workspace alongside your other content. No separate YouTube app, no algorithmic feed deciding what you see. When a creator you follow publishes something new, it's in Fabric, ready to be saved, transcribed, and searched.

Combine RSS with agents to automate processing: summarise new uploads from channels you follow, flag videos that mention topics you're tracking, or generate weekly digests of what's been published.

A library that compounds

Over months and years of saving the best YouTube content, the library grows into something more valuable than any playlist. The conference talk from last year connects to the tutorial from last month connects to the interview from last week. AI search finds connections across time. The explorer surfaces relationships between videos and other material. Smart organization tags and categorises by content without manual playlist maintenance.

The investment in saving and transcribing compounds. A question asked across a year of saved YouTube videos produces a richer answer than the same question across a month. See second brain for the full approach.

Connected to everything else you know

A YouTube video in the YouTube app is disconnected from your notes, your articles, your documents, and your work. In Fabric, the video transcript lives alongside every other content type. The conference talk connects to the paper it referenced. The tutorial connects to the project you're applying it to. The interview connects to your notes about the same topic.

This connectivity matters most for research, content planning, and studying, where YouTube content is one source among many and the value comes from connecting across all of them.

Who uses Fabric for YouTube videos

YouTube learning spans every domain. Students save educational videos and lecture recordings for studying. Developers save conference talks, tutorials, and technical deep-dives. Researchers transcribe and search academic talks and interviews. Writers and content creators collect references for content planning. Founders and indie hackers save strategy talks, product demos, and founder interviews. Educators curate video resources for teaching. Music creators save production tutorials and music theory content.

For the broader personal learning approach, see reading and learning and building a second brain. For capturing what you learn from podcasts, see podcast notes. For your own video recordings, see videos.

Get started

Start building a searchable library from the best of YouTube. Try Fabric free.

Save videos with the web clipper from your browser or share from the mobile app.

FAQs

Can Fabric transcribe YouTube videos?

Yes. Save a video and audio and video transcription generates a timestamped, searchable transcript.

Can I find a specific moment without scrubbing?

Yes. AI search reads every transcript and searches by meaning. Describe what was said and find the moment with a timestamp.

Can the AI summarise a long video?

Yes. The AI assistant can summarise any video, extract key takeaways, or answer specific questions about what was said.

Can the AI search across multiple videos?

Yes. Ask a question and the assistant draws from every relevant transcript. Find every talk where someone discussed a specific topic and get a synthesised answer.

Can I embed YouTube videos on the canvas?

Yes. The canvas supports live YouTube embeds. Videos are playable alongside images, notes, and other references.

Can I annotate video transcripts?

Yes. Annotations let you highlight and comment on transcripts. Your annotations are searchable across your library.

Can I subscribe to YouTube channels in Fabric?

Yes. RSS feeds bring new uploads into your workspace. Combine with agents for automated summaries.

Can I search across YouTube videos and my other content together?

Yes. Video transcripts are searchable alongside articles, notes, documents, and everything else. A concept from a YouTube talk and a related article are findable in the same search.

Can I write notes alongside a video?

Yes. Write in notes and docs alongside the transcript. Your observations and the video content are connected and searchable together.

Does the library get more useful over time?

Yes. Every video you save deepens what search and the AI can draw on. The investment in transcribing and saving compounds into a searchable knowledge base.

How is this different from YouTube's built-in search?

YouTube search finds videos on the platform. Fabric search finds moments within videos you've saved, by meaning, alongside your other content. The difference is between finding a video to watch and finding the insight within a video you've already watched, connected to everything else you know.

Are my saved videos private?

Yes. Your content is encrypted and only visible to you unless you choose to share it. Fabric uses AES-256 encryption and is CASA Tier 2 compliant. Your data is never used to train AI models.