Create AI Videos from Prompts with OpenAI, Script, TTS, and Pexels B-Roll Assembly

This n8n workflow automates the creation of short AI-generated videos. It takes a text prompt, uses OpenAI to generate a video script, converts the script to speech, and then assembles the video using Pexels stock footage.

What it does

This workflow streamlines the video creation process through the following steps:

1. Receives a Prompt: It starts by listening for an incoming webhook request containing a text prompt for the video.
2. Generates Video Script: Uses the OpenAI node to generate a video script based on the provided prompt.
3. Generates Audio from Script: Converts the generated script into an audio file using a Text-to-Speech (TTS) service (likely via a custom script executed by the Execute Command node).
4. Searches for B-Roll Footage: Queries Pexels for relevant stock video footage based on keywords extracted from the script or prompt.
5. Downloads Pexels Footage: Downloads the selected B-roll video clips from Pexels.
6. Assembles Video: Combines the generated audio and downloaded video clips into a final video file (likely using a custom script executed by the Execute Command node).
7. Uploads to Google Drive: Stores the final video file in a specified Google Drive folder.
8. Responds to Webhook: Provides a response back to the initial webhook, potentially including a link to the generated video.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• OpenAI API Key: For generating the video script.
• Pexels API Key: For searching and downloading stock video footage.
• Google Drive Account: For storing the generated videos.
• Google Drive Credentials: Configured in n8n for Google Drive access.
• Text-to-Speech (TTS) Service: This workflow appears to use a custom script via the Execute Command node for TTS. You will need to ensure the underlying system has the necessary TTS tools installed and configured (e.g., ffmpeg, a specific TTS engine, etc.).
• Video Editing/Assembly Tool: Similar to TTS, the video assembly is handled by a custom script via Execute Command. This implies the n8n host system needs video editing tools (e.g., ffmpeg) installed and configured to combine audio and video.

Setup/Usage

1. Import the Workflow: Download the JSON provided and import it into your n8n instance.
2. Configure Credentials:
   • Set up your OpenAI API Key credential in n8n.
   • Set up your Google Drive OAuth2 credential in n8n.
   • The Pexels API key and any custom script parameters will likely need to be configured within the respective HTTP Request and Execute Command nodes, or as environment variables if your scripts are designed to read them.
3. Configure Webhook:
   • Activate the Webhook node and copy its URL. This URL will be used to trigger the workflow.
4. Customize Execute Command Nodes:
   • Review the Execute Command nodes. These are critical for TTS and video assembly. You will need to ensure the commands are correct for your system's setup and that all necessary tools (like ffmpeg and your chosen TTS engine) are installed and accessible on the machine hosting n8n.
   • Adjust any paths or parameters within these commands to match your environment.
5. Test the Workflow:
   • Send a POST request to the Webhook URL with a JSON body containing your video prompt, for example:

     {
       "prompt": "A short video about the benefits of remote work, showing people collaborating happily from different locations."
     }
     

   • Monitor the execution in n8n to ensure all steps complete successfully.
   • Check your configured Google Drive folder for the generated video.