Repurpose YouTube Transcripts into Podcast Scripts using AI - n8n Workflow

Content Creators & Repurposers: Individuals who need to quickly convert long-form video content into clean, structured audio scripts.
Podcasters: Teams looking to rapidly draft episode scripts based on existing YouTube video discussions.
n8n Automation Engineers: Users seeking an advanced example of leveraging multiple external AI APIs (Dumpling AI, OpenAI) within a single, powerful n8n workflow.
Agencies: Organizations requiring a scalable solution for content transformation using n8n templates.

Manually converting lengthy video transcripts into usable podcast scripts is tedious, error-prone, and time-consuming. This specialized n8n workflow eliminates that burden entirely. It acts as a fully automated content repurposing engine.

The process begins whenever a new video is published to a monitored YouTube channel. The n8n trigger captures the video link, which is immediately sent to Dumpling AI—an external service used here as an API n8n node—to reliably extract the full transcript, including speaker differentiation.

The core intelligence of this n8n template lies in the subsequent step, where the raw transcript is fed into the powerful GPT-4o model. Using precise system instructions, this AI n8n node cleans the text by removing filler words, formats it professionally with labeled speakers, summarizes the key content, and extracts a title. Crucially, the n8n workflow forces the GPT-4o response to be in a structured JSON format. Finally, the structured output is stored neatly in an Airtable database, making it immediately available for review or further production steps. This complex n8n automation saves countless hours and ensures content consistency.

This highly efficient n8n workflow follows four major sequential steps, initiated by an n8n trigger:

1. YouTube Monitoring: The workflow starts with the 'Watch for New YouTube Video via RSS' n8n trigger. This node polls the specified YouTube channel RSS feed at regular intervals, activating the entire n8n workflow upon detecting a new video link.


2. Transcript Extraction: The video URL is passed to the 'Get Transcript from YouTube Video using Dumpling AI' HTTP Request n8n node. This node connects to the Dumpling AI API via a POST request, instructing it to quickly retrieve the full, detailed transcript for the video link supplied by the initial n8n trigger.


3. AI Transformation: The raw transcript then enters the 'Transform Transcript into Podcast Script using GPT-4o' n8n node. This is an OpenAI node utilizing GPT-4o. The n8n template provides specific instructions to clean the transcript (remove filler words, label speakers), generate a concise summary, and extract a title. It strictly enforces a JSON output structure for reliable downstream processing.


4. Data Storage: The final, cleaned, and structured data (title, summary, and cleaned transcript) is collected and routed to the 'Save Podcast Script and Metadata to Airtable' n8n node. This node creates a new record in the designated Airtable base, completing the automated content delivery cycle of this n8n workflow.

To deploy this powerful n8n workflow, follow these steps:

1. Import the n8n template: Copy the provided JSON data and paste it directly into your n8n instance using the 'New' -> 'Import from JSON' option.


2. Configure the RSS n8n trigger: In the 'Watch for New YouTube Video via RSS' n8n node, update the Feed URL parameter to the RSS feed corresponding to the YouTube channel you wish to monitor.


3. Set up Dumpling AI Credentials: The 'Get Transcript from YouTube Video' n8n node requires HTTP Header Authentication. You must create or select a credential set containing your Dumpling AI API key.


4. Set up OpenAI Credentials: Configure the 'Transform Transcript into Podcast Script' n8n node with your OpenAI API Key (using a credential linked to the GPT-4o model).


5. Set up Airtable Credentials and Target: In the 'Save Podcast Script and Metadata to Airtable' n8n node, provide your Airtable Personal Access Token credential. Then, specify the correct Base and Table where the resulting podcast scripts and metadata will be stored. Ensure your table contains columns matching the expected output fields: 'Title', 'summary', and 'podcast transcript'.


6. Activate: Once all credentials are set and configurations are saved, activate the n8n workflow to begin monitoring the YouTube feed.

This n8n workflow relies on several specialized n8n nodes:

RSS Feed Read Trigger (n8n trigger):
Function: Serves as the starting n8n trigger, monitoring an external RSS feed (your YouTube channel) for new items.
Key Configuration: Configured to poll the provided feedUrl for new video links.

HTTP Request (Dumpling AI):
Function: Executes an authenticated POST request to the Dumpling AI API to retrieve the complete, time-coded transcript of the new YouTube video identified by the initial n8n trigger.
Key Configuration: Uses a generic HTTP Header Auth credential. The JSON body dynamically inserts the {{ $json.link }} variable from the RSS feed as the videoUrl.

OpenAI (GPT-4o n8n node):
Function: Takes the raw transcript and uses the advanced GPT-4o model to perform complex content transformation, cleaning, summarization, and title extraction. This is a critical step in the n8n workflow.
Key Configuration: Uses the chatgpt-4o-latest model. Crucially, the node is set to JSON Output mode, ensuring the response is reliably structured with keys for title, summary, and cleaned_transcript.

Airtable n8n node:
Function: The final action of the n8n workflow, creating a new record in the specified Airtable table using the structured metadata and cleaned script generated by the OpenAI n8n node.
Key Configuration: Operation set to Create. Data mapping uses expression links (e.g., ={{ $json.message.content.title }}) to populate Airtable columns directly from the GPT-4o JSON output.