# How to find reusable clips in your YouTube back catalog

> Start by deciding whether the clip you want is something you said or something you showed.

Canonical URL: https://vivu.ai/guide/how-to-find-reusable-clips-in-my-youtube-back

Start by deciding whether the clip you want is something you said or something you showed. If you said it, search the transcripts: YouTube shows one per video, and you can download the subtitle files for your videos from YouTube Studio and search them all together. If it's something visual, like a b-roll shot, a reaction or a location, transcripts won't help, and you either scrub the footage or use a tool that searches the picture itself. Either way, cut from your original files, not from what's on YouTube.

## Decide what "reusable" means first

A back catalog holds a few different kinds of reusable clip, and each kind is found a different way. A line you said that works as a callback in a new video lives in the transcript. An explanation you gave once and don't want to film again also lives in the transcript, as long as you remember roughly how you phrased it. A drone shot of a city you visited, your face when something went wrong, or a product sitting on your desk usually lives only in the picture.

Write down three or four specific moments you want before you open anything. "Old footage I could reuse" gives you nothing to search for. "The time I explained why I switched microphones" does.

## Route 1: memory and your channel's own search

The search box on your channel page matches titles and descriptions. It's quick and it works when your titles say what the video is about. It stops working on older uploads with vague titles, and it can't point to a minute inside a video. Most people start here and outgrow it once the catalog passes the size they can hold in their head.

## Route 2: transcripts, searched together

Opening the transcript on each video and searching it one at a time works for a handful of videos and gets slow after that. Having transcripts isn't the problem. Searching across all of them at once is. The fix is to download the subtitle files from YouTube Studio into one folder and search that folder with any text search on your computer. Each line in the file carries a timestamp, so a hit tells you which video and roughly where.

This has limits. Automatic captions mishear names and jargon, so search for the words around the one you care about too. It only finds words you remember, and it finds nothing that wasn't said out loud. There's more on this approach in [searching video by what was said](https://vivu.ai/guide/search-video-by-spoken-words).

## Route 3: a log, from now on

A spreadsheet row per upload, with a few notes on moments worth reusing, costs a couple of minutes per video. It does nothing for the catalog you already have, and it only works if you keep at it. For the catalog going forward, [keeping footage findable over years](https://vivu.ai/guide/how-creators-keep-track-of-footage-across-years) covers folder and naming habits that hold up.

## Route 4: upload a batch to a video search tool

The last route is to upload the original files to a tool that indexes what's in the video, then describe the moment in plain language. What comes back is a set of segments, often from several different videos, which is the point: one question can pull together moments you'd never have connected yourself. You still judge each one. Some results will be only indirectly related to what you asked, and deciding whether they're usable is your call. The cost is the upload and the fact that indexing counts against the tool's limits, so start with the videos you most want to reuse from.

With [Vivu](https://vivu.ai/platform), you upload the video files for those episodes into a project, they're indexed once, and a query like "me explaining why I switched microphones" comes back as time ranges across different videos, each with a short reason, which you open, check, and export as original clips for your own edit. It has a free tier and a paid tier, with the limits listed on its pricing page (`/pricing`), which is another reason to begin with a small batch and not the whole channel.

## When you don't need any of this

If your catalog is small enough that you remember what's in it, you already have your answer. If every clip you reuse is something you said, downloaded transcripts in a folder will cover most of it for free. And if you're planning a whole compilation video, not grabbing single moments, [planning a compilation from old uploads](https://vivu.ai/guide/how-to-plan-a-compilation-video-from-old-uploads) is a better fit for the job.

Your route depends on the moment you're chasing. If you can remember roughly what you said, the transcripts will get you there. If you can only picture what it looked like, you'll be scrubbing or searching the video itself.

## FAQ

### Can I search all my YouTube videos at once for something I said?

The most reliable way is to download the subtitle files for your videos from YouTube Studio and search them together on your computer. YouTube's own search is built to find videos, not to jump to a line inside one, and the transcript panel on a video only covers that one video.

Each subtitle file includes timestamps, so a match shows you both the video and roughly where in it the line is.

### Can I reuse clips from my old videos without copyright problems?

Your own footage is yours, but old videos can contain things that aren't. Music, licensed stock clips and guest appearances may have come with terms that covered the original video and not a new one. Check those before you re-cut, because a music claim can follow a clip into a new upload.

### Should I use my original files or download my videos from YouTube?

Use the originals whenever you have them. A download from YouTube Studio is a compressed copy, which is fine for finding the moment but gives you a worse clip to cut. Once you've found the moment, go back to the original file and export from that.

### How do I find b-roll in old videos when nothing was said over it?

Transcripts can't find a shot that has no speech over it. Your options are scrubbing through footage at high speed, checking old project files for where you placed the shot before, or using a video search tool that indexes the picture and lets you describe the shot you're after.