n8n Video Voiceover & Subtitle Automation with OpenAI Whisper, TTS, and FFmpeg

This n8n workflow automates the process of generating voiceovers and subtitles for video files using OpenAI's Whisper for transcription, OpenAI's Text-to-Speech (TTS) for voiceover generation, and FFmpeg for video processing. It provides a flexible way to handle video files from various sources and apply AI-powered audio and text enhancements.

Description

This workflow streamlines the complex task of adding voiceovers and subtitles to videos. It's designed to be highly adaptable, allowing you to trigger the process manually, via a form submission, or by integrating with file storage services like Google Drive or FTP. Once a video file is provided, the workflow leverages advanced AI models to transcribe the audio, generate a new voiceover, and prepare the necessary data for subtitle creation, ultimately enabling automated video localization and accessibility.

What it does

1. Triggers: The workflow can be initiated in several ways:
   • Manually: By clicking 'Execute workflow' in the n8n editor.
   • On Form Submission: By submitting data through a configured n8n form.
   • From Google Drive: By detecting new or updated files in a specified Google Drive folder.
   • From FTP: By monitoring an FTP server for new or updated files.
   • Via SSH: Potentially by executing commands on a remote server to retrieve file information or trigger processes.
   • HTTP Request: By receiving an HTTP request, likely containing video file details or a URL.
2. Edit Fields (Set): Prepares and structures the incoming data for subsequent processing, ensuring consistency and extracting relevant information like file paths or URLs.
3. Merge: Combines data from different branches or previous steps, ensuring all necessary information (e.g., original video details, transcription results, TTS audio) is available for the next stages.
4. OpenAI Integration:
   • Whisper (Transcription): Transcribes the audio track of the input video file into text.
   • Text-to-Speech (TTS) (Voiceover): Generates a new audio voiceover from the transcribed text, potentially in a different language or with a different voice.
5. HTTP Request: Likely used to interact with an external FFmpeg service or API to perform video and audio manipulation, such as:
   • Merging the generated voiceover audio with the original video.
   • Embedding subtitles (SRT/VTT) into the video.
   • Converting video formats or adjusting properties.
6. Code: Executes custom JavaScript code to handle complex data transformations, manipulate file paths, or prepare data for FFmpeg commands based on the AI outputs.
7. Sticky Note: Provides documentation or notes within the workflow for clarity and understanding of specific steps.

Prerequisites/Requirements

• n8n Instance: A running n8n instance (cloud or self-hosted).
• OpenAI API Key: An API key for OpenAI services (Whisper for transcription, TTS for voice generation).
• FFmpeg Service/API: Access to an FFmpeg installation or a service that can execute FFmpeg commands for video processing. This workflow uses an HTTP Request node, implying an external service.
• Google Drive Account (Optional): If using the Google Drive trigger, an authenticated Google Drive credential in n8n.
• FTP Server Access (Optional): If using the FTP trigger, FTP server credentials configured in n8n.
• SSH Access (Optional): If using the SSH node, SSH credentials configured in n8n.

Setup/Usage

1. Import the Workflow: Download the JSON provided and import it into your n8n instance.
2. Configure Credentials:
   • OpenAI: Set up your OpenAI API key as a credential in n8n and link it to the "OpenAI" node.
   • Google Drive (Optional): If you plan to use the Google Drive trigger, configure your Google Drive OAuth2 credentials.
   • FTP (Optional): If you plan to use the FTP trigger, set up your FTP credentials.
   • SSH (Optional): If you plan to use the SSH node, configure your SSH credentials.
3. Configure Trigger Node:
   • Choose your preferred trigger (Manual, Form Submission, Google Drive, FTP, SSH, HTTP Request).
   • If using the "On form submission" trigger, customize the form fields as needed.
   • If using Google Drive or FTP, specify the folder/path to monitor.
   • If using "HTTP Request", configure the endpoint and expected payload.
4. Customize "Edit Fields (Set)" and "Code" Nodes: Adjust these nodes to match the exact structure of your input data and the desired output for FFmpeg commands. The "Code" node will likely contain logic for generating FFmpeg commands based on the transcription and TTS results.
5. Configure HTTP Request (FFmpeg): Update the "HTTP Request" node to point to your FFmpeg service/API endpoint and configure the request body with the necessary FFmpeg commands generated by the "Code" node.
6. Activate the Workflow: Once configured, activate the workflow to enable it to run automatically based on your chosen trigger.

This workflow provides a robust framework for automating video localization and accessibility, significantly reducing manual effort in creating voiceovers and subtitles.