n8n Workflow: Generate and Translate Video Subtitles

This n8n workflow provides a framework for generating and translating video subtitles. It leverages local command execution, HTTP requests, and file operations, suggesting an integration with external tools like OpenAI Whisper for transcription and LibreTranslate for translation.

What it does

This workflow outlines a process for:

1. Triggering the process: Initiated by a webhook, likely receiving a video file or a link to one.
2. Executing Local Commands: Runs shell commands, which would typically involve:
   • Downloading the video (if a URL is provided).
   • Processing the video with a tool like OpenAI Whisper to generate initial subtitles (e.g., in SRT format).
   • Potentially preparing the subtitle file for translation.
3. Reading Subtitle File: Reads the generated subtitle file (e.g., .srt or .vtt) from the local filesystem.
4. Translating Subtitles: Sends the subtitle content via an HTTP Request to a translation service (e.g., a self-hosted LibreTranslate instance or another translation API).
5. Writing Translated Subtitles: Saves the translated subtitle content to a new file on the local filesystem.
6. Merging Data: Combines the original and translated subtitle data, potentially for further processing or sending.
7. Emailing Results: Sends an email via Gmail, presumably containing the original and/or translated subtitle files, or a notification about the completion of the process.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: An n8n instance capable of executing shell commands.
• OpenAI Whisper (or similar ASR tool): This workflow is designed to integrate with a local installation of Whisper for speech-to-text transcription.
• LibreTranslate (or similar translation service): This workflow expects an endpoint for a translation service. This could be a self-hosted LibreTranslate instance or another translation API.
• Gmail Account: For sending email notifications or results. You'll need to configure Gmail credentials in n8n.
• Video Files: The video files you wish to process. The workflow likely expects either a direct file upload via the webhook or a URL to a video.

Setup/Usage

1. Import the Workflow: Import the provided JSON into your n8n instance.
2. Configure Webhook:
   • Activate the "Webhook" node and copy its URL. This URL will be used to trigger the workflow.
   • Configure the webhook to receive the video file or URL.
3. Configure Execute Command Node:
   • Modify the "Execute Command" node to include the specific commands for downloading videos, running OpenAI Whisper, and any other necessary shell scripts.
   • Ensure the n8n user has the necessary permissions to execute these commands and access local files.
4. Configure HTTP Request Node:
   • Update the "HTTP Request" node with the endpoint of your chosen translation service (e.g., LibreTranslate API URL).
   • Configure the request body to send the subtitle text for translation.
5. Configure Read/Write Binary File Nodes:
   • Adjust the file paths in the "Read Binary File" and "Write Binary File" nodes to match your desired storage locations for subtitle files.
6. Configure Gmail Node:
   • Set up your Gmail credentials in n8n.
   • Configure the "Gmail" node to send emails with the desired subject, body, and attachments (e.g., the generated and translated subtitle files).
7. Activate the Workflow: Once configured, activate the workflow. You can then trigger it by sending a request to the Webhook URL.