How to turn a YouTube transcript into a LinkedIn post — for free
A step-by-step guide to extracting a YouTube transcript and turning it into a properly-voiced LinkedIn post, including the free tools and methods that actually work.
YouTube's caption system is genuinely useful for content creators, and most people don't use it to its full potential. If you've made a YouTube video, you've also — without knowing it — made the raw material for multiple LinkedIn posts. Here's how to get from one to the other without spending any money on the first generation.
Step 1: Get the transcript
For any public YouTube video with captions enabled, you can access the full transcript directly on YouTube:
- Open the video on YouTube.
- Click the three-dot menu (⋯) below the video.
- Select "Show transcript."
- A panel opens on the right. Copy all the text.
Most videos with English-language audio have auto-generated captions. If the transcript panel doesn't appear, captions may be disabled by the creator or the video may be in a language without auto-captions. In that case, you can use a tool that downloads the audio and transcribes it directly — Verbatrum does this automatically for a small credit cost.
Step 2: Clean the transcript (optional but useful)
Auto-generated transcripts have two main problems: no punctuation, and speaking tics ("um," "you know," "like"). You don't need to clean the whole thing — just skim for the 3–5 minutes of content that's most relevant to the post you want to write, and clean that section.
You're looking for the specific moment where you made the point that matters. Not the introduction, not the recap — the actual argument, example, or insight.
Step 3: Extract the core insight, not the summary
Before writing anything, reduce what you found in step 2 to a single sentence: "The main thing I was actually saying was ___." This is harder than it sounds. Most transcripts contain three good ideas buried under a lot of setup.
The LinkedIn post should be built around that one sentence, not around a summary of everything you covered. A summary is low-value because it tells people what the video was about. The extracted insight tells them something they can use without watching the video.
Step 4: Rewrite it in your voice
This is where most DIY repurposing breaks down. The instinct is to paste the transcript into ChatGPT and say "write a LinkedIn post about this." The output will be grammatically correct, structurally sensible, and completely interchangeable with a thousand other posts.
The better approach is to write the post yourself, using the transcript only as a source for the facts and examples — not as a template to rephrase. If that sounds like a lot of work, the middle path is a tool that's been trained on your own past posts so it can write in your voice rather than the average LinkedIn voice. Verbatrum does this: paste the YouTube link, it extracts the transcript and the key points, and the rewrite uses a fingerprint extracted from your own writing. The Voice Match Score tells you how close the output got.
What a good result looks like
A well-converted YouTube-to-LinkedIn post should:
- Open with a hook that could only have come from that specific video — not a generic claim about the topic in general.
- Use an example or detail from the video as the evidence for the hook. Specificity is the signal that this isn't just an AI opinion.
- Close with something that makes the reader want to do something, not "what are your thoughts?" — a recommendation, a question with a real answer, or a single clear takeaway.
- Sound like you, not like a press release about you.
Free tools summary
- YouTube's built-in transcript viewer — free, works for any video with captions enabled.
- Verbatrum's YouTube → LinkedIn tool — free for the first generation (no account required). Uses captions when available, audio transcription as a fallback. Voice matching requires creating a free account and pasting a few examples of your writing.
The full pipeline — transcript extraction, key point distillation, voice-matched rewrite — takes about three minutes. The output is still a starting point, not a finished post. But it's a much better starting point than a blank page.