How to get YouTube transcripts in n8n, Make and Zapier

Checked on 2026-11-18.

Automation tools run in the cloud, and that is where the usual free ways of reading YouTube transcripts stop working: YouTube refuses most requests from data-centre addresses. An HTTP API avoids the problem, and every automation tool has a step for calling one. Here are the settings for the three most used tools.

The request, whatever the tool

It is one GET request with your API key in a header. You get a key, with 100 free credits, in the dashboard.

SettingValue
MethodGET
URLhttps://pipefetch.com/v1/youtube/transcript
Query parametervideo = the video's link or ID
Headerx-api-key = your key

The answer is JSON. The transcript is in data.text, and the video's title and channel are in data.title and data.channel.name.

n8n

Add an HTTP Request node.

  • Method: GET. URL: https://pipefetch.com/v1/youtube/transcript
  • Authentication: choose the generic credential type, and Header Auth as its kind. Create a credential with the name x-api-key and your key as the value. This keeps the key out of the workflow itself.
  • Send Query Parameters: on. Add video, and as its value an expression that points at the link from an earlier node, for example {{ $json.url }}.

In the next node the transcript is {{ $json.data.text }}. A video without captions answers with status 404, which n8n treats as a failed node by default. To let the workflow continue, open the node's Options, add Response and turn on Never Error; then check in an IF node whether {{ $json.data }} exists.

Make

Add the HTTP app and choose Make a request.

  • URL: https://pipefetch.com/v1/youtube/transcript. Method: GET.
  • Headers: add an item with the name x-api-key and your key as the value.
  • Query String: add an item with the name video and map the video link from an earlier module as its value.
  • Parse response: Yes, so the fields of the answer can be mapped in later modules.

The transcript is then available as data.text. For videos without captions, turn off the option that treats error statuses as errors, or add an error handler to the module, so one such video does not stop the scenario.

Zapier

Add an action with Webhooks by Zapier and the event GET. Webhooks is one of Zapier's premium apps, so it needs a paid Zapier plan.

  • URL: https://pipefetch.com/v1/youtube/transcript
  • Query String Params: video on the left, the video link from an earlier step on the right.
  • Headers: x-api-key on the left, your key on the right.

Zapier flattens the answer, so the transcript appears in later steps as a field named Data Text.

What people build with it

  • New video to summary. Watch a channel's uploads, fetch each new video's transcript and hand it to an AI step that writes a summary, a newsletter paragraph or a social post.
  • Videos into a knowledge base. Store transcripts in a database or a document tool, with the title and link, so they can be searched.
  • Research lists. Take video links from a spreadsheet column and write the transcripts back next to them.

For the first one, the channel's latest uploads come from a second endpoint, https://pipefetch.com/v1/youtube/channel/videos, with the parameter channel set to a handle such as @mkbhd. It is set up the same way.

What it costs

Each transcript is 1 credit, and a video without captions costs nothing. A workflow that handles 20 new videos a day uses about 600 credits a month. Every request is fetched live and takes a few seconds, which automation tools handle without any special setting.

Try it with 100 free credits. No card, no Google Cloud project, no OAuth. Get an API key or read the docs.