Summarize a Copied YouTube Transcript on Mac

YouTube transcripts are long. A 20-minute video can produce 3,000+ words of raw text with no paragraphs, no headings, and timestamps scattered everywhere. If you just want the three or four points that matter, reading the whole thing defeats the purpose. Here's how to copy a transcript and summarize it directly on your Mac without uploading anything to a website.

Get the transcript onto your clipboard

YouTube exposes transcripts under the video. Click the ...more below the video title, choose Show transcript, then select the text in the transcript panel and copy it with Cmd+C. If you prefer, toggle off timestamps in the transcript panel first so you copy clean sentences.

The transcript now lives on your clipboard. The problem: it's a wall of text, and pasting it into a note doesn't make it shorter.

Summarize it with an AI clipboard manager

ClipHistory keeps your last 150 unpinned clips and lets you run AI transforms on any of them. Open the clip list with Cmd+Shift+V, find the transcript you just copied, and run the Summarize transform.

Because ClipHistory uses your own API key, you choose the model behind the summary. You can connect Anthropic, OpenAI, DeepSeek, Google, or a custom endpoint. The request goes straight from your Mac to the provider you configured — there's no ClipHistory cloud in the middle, no account to create, and no copy of your transcript stored on a server.

Why your own API key matters here

Transcripts can contain a lot of context you didn't write — interviews, customer calls recorded as video, internal training. Routing that through a third-party SaaS summarizer means trusting their retention policy. With ClipHistory the data path is just: your clipboard → your API provider → back to your clipboard. Nothing is persisted beyond your local clip history, which lives on your machine.

Tighten the result

A first-pass summary is often still longer than you want. Two practical moves:

• Re-run Summarize on the summary. Each pass compresses further. Two passes usually gets a 3,000-word transcript down to a scannable paragraph.
• Switch to Rewrite and ask for a specific shape — a bulleted list of takeaways, or a single tweet-length line. The Rewrite transform is built for changing form, not just length.

Pin the summaries you reuse

If you're summarizing a series of videos for research, pin the good summaries. Pinned clips are unlimited and never get pushed out by the 150-clip rolling window, so your distilled notes stay put while raw transcripts cycle through.

A repeatable workflow

1. Open the YouTube transcript and Cmd+C the text.
2. Hit Cmd+Shift+V to open ClipHistory.
3. Run Summarize on the transcript clip.
4. Re-run or switch to Rewrite if you need a tighter shape.
5. Pin or paste the result wherever you're taking notes.

The whole loop takes a few keystrokes and never leaves your Mac except for the AI call you explicitly configured.

What this is good for

• Turning long tutorials into a checklist before you follow along.
• Pulling the argument out of a video essay without watching at 2x.
• Capturing the decisions from a recorded meeting that was posted as a video.
• Pre-reading a conference talk so you can decide whether the full video is worth your time.
• Building a study guide from a lecture series, one summarized transcript at a time.

Because the transform runs on whatever is in your clip history, you're not limited to YouTube. Any transcript you can copy — from a transcription app, a podcast tool, or a meeting recorder — can be summarized the same way.

Clean the transcript first for better results

Raw transcripts are noisy. Auto-generated captions repeat filler words, drop punctuation, and sometimes splice the speaker's "um" and "you know" right into the sentence. Feeding that directly to a summarizer still works, but you'll get a tighter result if you clean it up first.

ClipHistory includes a Clean transform for exactly this. Run Clean on the transcript clip to strip stray line breaks, collapse the filler, and normalize the text into readable sentences. Then run Summarize on the cleaned version. Because each transform's output becomes a new clip, chaining them is just two keystrokes apart — Clean, then Summarize, with no copy-paste detour in between.

This Clean-then-Summarize pattern is the single biggest quality improvement for transcript work. A model spends less of its attention untangling formatting noise and more of it on the actual content, so the summary tends to be sharper and shorter.

Where everything runs

It's worth being explicit about the data path, because transcripts can be sensitive. ClipHistory is signed and notarized by Apple and runs as a universal binary on both Apple Silicon and Intel Macs under macOS 12 or later. Your clip history is stored on your machine. The only time anything leaves your Mac is the AI request you deliberately trigger, and that goes straight to the provider whose key you configured — there is no ClipHistory account, no telemetry of your clips, and no cloud copy of your transcript. You open it all with the global shortcut Cmd+Shift+V, so the entire workflow lives a keystroke away from whatever app you're in.

Ready to put AI to work on your clipboard? Get ClipHistory for macOS ($19.99) — a one-time payment, 12-month license, no auto-renewal. Signed and notarized by Apple, runs on macOS 12+, and everything stays local on your Mac.