Generate Social Media Content from Video Transcripts with Gemini AI & Airtable

This n8n workflow automates the process of generating social media content from video transcripts using Google Gemini AI and then storing the generated content in Airtable. It's designed to streamline content creation for video producers, marketers, or anyone looking to repurpose video content efficiently.

Description

This workflow simplifies the task of transforming raw video transcripts into engaging social media posts. By leveraging the power of Google Gemini's AI agent and structured output parsing, it extracts key information and generates various social media formats (e.g., short posts, tweets, LinkedIn updates) from a given transcript. The generated content is then neatly organized and stored in an Airtable base, ready for review and scheduling.

What it does

1. Triggers Manually or via Webhook: The workflow can be initiated manually with a click or by sending a POST request to a defined webhook URL.
2. Retrieves Transcript from Airtable: It fetches a video transcript from a specified Airtable base, likely based on a trigger or a predefined record ID.
3. Prepares Data: An "Edit Fields (Set)" node processes the incoming data, likely extracting the relevant transcript text and preparing it for the AI agent.
4. Generates Social Media Content with AI: An "AI Agent" node, powered by the "Google Gemini Chat Model" and guided by a "Structured Output Parser", takes the video transcript and generates social media content in a structured JSON format. This typically involves defining specific prompts for different social media platforms or content types.
5. Stores Content in Airtable: The generated social media content, now in a structured format, is then sent back to Airtable to be stored in a designated table, allowing for easy management and publishing.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance (cloud or self-hosted).
• Airtable Account: An Airtable account with a base and table set up to store video transcripts and the generated social media content. You will need an API key and the Base ID/Table Name.
• Google Gemini API Key: Access to the Google Gemini API (likely via Google Cloud Platform) to use the Google Gemini Chat Model.
• Google Drive (Optional): While a Google Drive node is present, its specific role in this workflow's current configuration is unclear from the provided JSON. It might be for fetching transcripts from Drive or storing generated content/metadata. If it's not actively used, it can be removed.

Setup/Usage

1. Import the workflow: Download the JSON and import it into your n8n instance.
2. Configure Credentials:
   • Airtable: Set up your Airtable credentials, linking them to the "Airtable" nodes. Ensure you specify the correct Base ID and Table Names for both reading transcripts and writing generated content.
   • Google Gemini Chat Model: Configure your Google Gemini API key as a credential for the "Google Gemini Chat Model" node.
3. Customize AI Agent Prompts:
   • Open the "AI Agent" node. You will need to define the specific prompts and instructions for Gemini to generate the desired social media content. This is where you tell the AI what kind of posts you want (e.g., "Generate 3 tweets, 1 LinkedIn post, and a short Instagram caption from the following transcript, focusing on key takeaways.").
   • Adjust the "Structured Output Parser" to match the expected JSON format of your AI agent's output.
4. Configure Webhook (if using): If you plan to trigger the workflow via a webhook, copy the webhook URL from the "Webhook" node and configure your external system to send a POST request to it.
5. Test the workflow: Run the workflow manually to ensure all nodes are configured correctly and the content is generated and stored as expected.
6. Activate the workflow: Once tested, activate the workflow to enable automatic execution (if using the webhook trigger).