YouTube to Google Docs, transcripts ready to publish
You grab a YouTube link, thinking “I’ll turn this into a blog post later.” Then you remember the messy part: finding a transcript, cleaning timestamps, fixing speaker breaks, and pasting everything into a doc that still looks terrible.
This is exactly where YouTube transcript automation pays off. Content marketers feel it when deadlines pile up. Educators creating study notes feel it too. And agency teams repurposing client webinars? They’re stuck in the same loop.
This n8n workflow turns a YouTube URL into a clean Google Doc transcript, with optional summary and translation. You’ll see what it does, what you need, and how the pieces fit together.
How This Automation Works
The full n8n workflow, from trigger to final output:
n8n Workflow Template: YouTube to Google Docs, transcripts ready to publish
flowchart LR
subgraph sg0["Summarize and Transl Flow"]
direction LR
n0@{ icon: "mdi:cog", form: "rounded", label: "Append Summary and Translation", pos: "b", h: 48 }
n1@{ icon: "mdi:cog", form: "rounded", label: "Create Output Document", pos: "b", h: 48 }
n2@{ icon: "mdi:robot", form: "rounded", label: "Summarize and Translate Text", pos: "b", h: 48 }
n3["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Combine Transcription Content"]
n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Transcribe YouTube Video"]
n5["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Format Webhook Input"]
n6["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/webhook.dark.svg' width='40' height='40' /></div><br/>Trigger YouTube Processing R.."]
n7@{ icon: "mdi:brain", form: "rounded", label: "Google Gemini To Summarize a..", pos: "b", h: 48 }
n5 --> n4
n1 --> n0
n4 --> n3
n2 --> n1
n3 --> n2
n6 --> n5
n7 -.-> n2
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n2 ai
class n7 aiModel
class n4,n6 api
class n3,n5 code
classDef customIcon fill:none,stroke:none
class n3,n4,n5,n6 customIcon
The Problem: Turning YouTube into usable text is annoyingly manual
Getting value from a video shouldn’t require a mini production process, but it usually does. You track down a transcript (or generate one), then spend ages removing junk: timestamps, weird line breaks, “um” spam, and repeated phrases. After that, you still have to format it into something shareable, and if you need a translated version, you’re basically doing the whole thing twice. It’s slow, mentally draining, and easy to abandon halfway through, which means good content ideas just sit there.
The friction compounds. Here’s where it breaks down in real life.
- Copying transcript chunks from different screens leads to missing lines and awkward gaps.
- Manual cleanup takes about an hour per long video, even if you’re fast.
- Summaries and translations become “nice-to-have,” so they never get done.
- Teams can’t standardize output, so every doc looks different and needs extra review.
The Solution: Send a YouTube link, receive a formatted Google Doc
This workflow starts with a simple webhook request that includes three inputs: the YouTube URL, a target language, and a toggle that controls whether you want a summary. n8n parses that payload, extracts a usable video ID, then sends the video to a transcription API (Supadata, in the default setup) via HTTP Request. Once the transcript comes back in segments, the workflow merges those parts into one readable body of text. Next, it uses a Gemini chat model through n8n’s LLM Chain to summarize and translate the transcript when you’ve enabled that option. Finally, it creates a Google Doc named from the video ID and language, then inserts the finished content so it’s ready for review or publishing.
The flow begins when you POST a URL to the webhook. The transcript is fetched and stitched into one clean text block. Gemini optionally adds a summary and a translated version. Then Google Docs gets a new document with everything inserted in the right place.
What You Get: Automation vs. Results
| What This Workflow Automates | Results You’ll Get |
|---|---|
|
|
Example: What This Looks Like
Say you repurpose three YouTube videos a week into written content. Manually, you might spend about 40 minutes pulling and cleaning each transcript, plus another 20 minutes making a summary and translation, so that’s roughly 3 hours weekly. With this workflow, you submit each URL in under a minute and wait for processing (often about 10 minutes per video). You still review the final doc, but you’re starting from “ready to edit,” not raw mess.
What You’ll Need
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Supadata (or similar transcript API) to fetch full video transcripts.
- Google Docs to create and fill the output document.
- Google Gemini / Vertex AI access (enable it in Google Cloud) for summary and translation.
Skill level: Intermediate. You’ll connect accounts, add API keys, and test a webhook call, but you won’t be writing an app.
Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).
How It Works
A webhook receives your request. You send a POST with youtube_url, language, and enable_summary. That’s it. No logging into five tools first.
The workflow cleans up the input. A code step extracts the video ID and normalizes the values so downstream tools don’t choke on odd URL formats or missing fields.
Transcript + AI processing happens in the middle. n8n calls the transcription API with HTTP Request, then merges the returned segments into one body of text. If the summary toggle is enabled, the LLM Chain sends the transcript to Gemini to generate a summary and translation in your chosen language.
Google Docs is created and filled. The workflow creates a new doc (named using the video ID and language), then appends the summary/translation content so the document is ready to share or edit.
You can easily modify the language handling to create one doc per language based on your needs. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Webhook Trigger
Set up the inbound webhook that receives YouTube processing requests.
- Add the Incoming YouTube Hook node.
- Set HTTP Method to
POST. - Set Path to
youtube. - Ensure the webhook is connected to Parse Webhook Payload.
Step 2: Parse and Fetch Transcript Data
Extract the video ID and call the transcript API.
- In Parse Webhook Payload, keep the JavaScript Code as provided to extract
videoId,language, andenable_summaryfrom the request body. - Connect Parse Webhook Payload to Fetch Video Transcript.
- In Fetch Video Transcript, set URL to
https://api.supadata.ai/v1/youtube/transcript?. - Enable Send Query and add a query parameter named
videoIdwith value={{ $json.videoId }}. - Enable Send Headers and add header
x-api-keywith your API key value.
⚠️ Common Pitfall: The x-api-key header in Fetch Video Transcript is empty by default—your transcript API call will fail unless you add a valid key.
Step 3: Set Up Transcript Processing and AI Generation
Merge transcript segments and generate translation/summary with the AI chain.
- Connect Fetch Video Transcript to Merge Transcript Segments.
- In Merge Transcript Segments, keep the JavaScript Code that merges
$json["content"]intomergedTranscript. - Connect Merge Transcript Segments to Generate Summary & Translation.
- In Generate Summary & Translation, set Text to the provided prompt, including the expressions
{{ $('Merge Transcript Segments').item.json.language }}and{{ $('Merge Transcript Segments').item.json.mergedTranscript }}. - Connect Gemini Chat Model to Generate Summary & Translation as the language model.
- Credential Required: Connect your googlePalmApi credentials in Gemini Chat Model (the AI model credentials are added on the model node, not the chain).
Step 4: Configure Google Docs Output
Create a new document and insert the AI output.
- Connect Generate Summary & Translation to Create Results Doc.
- In Create Results Doc, set Title to
={{ $('Parse Webhook Payload').item.json.videoId }}_{{ $('Parse Webhook Payload').item.json.language }}. - Set Folder ID to
1FM-kd1_Xi2ikKpK7xAULFWYc_qVH2wwc. - Credential Required: Connect your googleDocsOAuth2Api credentials in Create Results Doc.
- Connect Create Results Doc to Insert Summary & Translation.
- In Insert Summary & Translation, set Operation to
updateand Document URL to={{ $json.id }}. - Set the insert Text action to
={{ $('Generate Summary & Translation').item.json.text }}. - Credential Required: Connect your googleDocsOAuth2Api credentials in Insert Summary & Translation.
Step 5: Test and Activate Your Workflow
Validate that the webhook, transcript fetch, AI processing, and document output all work end to end.
- Click Execute Workflow and send a POST request to the Incoming YouTube Hook URL with a body containing
youtube_url,language, andenable_summary. - Confirm Fetch Video Transcript returns transcript content and Merge Transcript Segments outputs
mergedTranscript. - Verify Generate Summary & Translation returns valid JSON with
translationandsummary. - Check that Create Results Doc created a Google Doc and Insert Summary & Translation inserted the content.
- Once successful, toggle the workflow to Active for production use.
Common Gotchas
- Supadata (or your transcription API) credentials can expire or need specific permissions. If things break, check your API dashboard and the Authorization header in the HTTP Request node first.
- If you’re using Wait nodes or external rendering, processing times vary. Bump up the wait duration if downstream nodes fail on empty responses.
- Gemini prompts in the LLM Chain start generic. Add your preferred format (headings, tone, “keep quotes,” bullet style) early or you will be editing outputs forever.
Frequently Asked Questions
About 30 minutes if your Google account and API keys are ready.
No. You’ll mostly connect accounts and paste in an API key. The only “technical” part is sending a test webhook request, which you can copy from the implementation guide.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in transcription API costs and Gemini/Vertex usage for summarization and translation.
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
Yes, but you’ll tweak the flow a bit. The easiest approach is to pass an array of languages, then loop over them and run the “Generate Summary & Translation” step once per language. From there, create one Google Doc per language (or append separate sections into a single doc). Most people also adjust the doc title expression so the language is obvious at a glance.
Most of the time it’s OAuth permissions or an expired Google login in n8n. Reconnect the Google Docs credential and confirm it has access to create and edit documents in the right Google account. Also check where the doc is being created if your organization restricts Drive locations. If it works in tests but fails later, you may be hitting a shared-drive policy or token refresh issue.
On n8n Cloud Starter, you can usually run a few thousand executions per month, and self-hosting has no fixed execution cap (it depends on your server). Practically, your transcription API limits and video length matter more than n8n does.
Often, yes. This workflow benefits from code steps, conditional logic (the summary toggle), and an LLM chain, which can get pricey or awkward in other tools. n8n is also easier to self-host if you want predictable costs. That said, if you only need “URL in, text out” with zero customization, Zapier or Make can be quicker to click together. Talk to an automation expert if you’re unsure.
Once this is running, every good video instantly becomes written material you can actually use. Set it up once, then move on to the work that needs your taste and judgment.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
