Automating WordPress with Make.com: From Post to Cloud Voiceover

📦 GitHub: paoloronco/makecom-wordpress-ttsvoiceover-notiondb
📣 Publishing on ReadyTensor: GCP TTS on WordPress | WordPress to Voice

In the world of technology and cybersecurity, automate it's not just a question of efficiency, but also of precision and safety.
As Cyber Security Analyst with a strong passion for the tech automation, I developed an advanced workflow on Make.com which allows me to intelligently manage the entire life cycle of the content published on WordPress, transforming them into Multi-format resources and stored neatly across multiple platforms.

In this article I will show you How does my workflow work?, what makes it effective and how it could be adapted to other contexts, always in a key scalable and secure.

🎯 Workflow Goal

Automate WordPress post management in two directions:

  • Structured storage and notification (via Notion and Email),
  • Automatic generation of audio voiceovers and dynamically update the post on the site.

🔁 Process Overview

Step 1 – Watch Post from WordPress

The flow is activated every time a new post is published (or updated) on WordPress, thanks to the “Watch Post via API” module.

Step 2 – Router

The router splits the flow into two independent branches, which perform parallel operations on the same content.

Branch 1 – Storage in Notion Database

  1. The content of the post is added to a Notion table, useful for tracking, future consultation and content management in knowledge base format.
  2. At the end, a Email notification which confirms that the post has been inserted correctly.

Useful for: structured backup, content history, internal documentation.

Branch 2 – VoiceOver Audio Generation

1. Text Parsing and Optimization

  • The post content is converted from HTML to clean text.
  • The text is then Optimized and reformatted using a model of OpenAI GPT, to improve its readability and prepare it for speech synthesis.

2. Upload to Google Cloud Storage

The text is loaded in format .txt on a private bucket on GCP.

3. Control with Custom JS

A script in Custom JavaScript check the size of the generated file:

  • If it is less than 5000 bytes, the text is short → a standard TTS synthesis is used.
  • If it is greater, an alternative path is activated for long files.

The JS code returns a boolean value which is handled by a second router.

4. Router: Choosing between Short TTS and Long TTS

If text is < 5000 bytes:

  1. The file is sent to the form Google Cloud Text-to-Speech to generate audio .wav.
  2. The audio is uploaded to Google Drive.
  3. It is downloaded via an HTTP module and then re-uploaded to a public GCP bucket.
  4. The updated content of the post on WordPress is retrieved.
  5. The new audio comes integrated within the post with a custom HTML block (see example below).
  6. Final confirmation email.

If text is > 5000 bytes:

  1. A applies Advanced Text Parser which removes special characters and further cleans up the content.
  2. An HTTP module activates a Cloud Run Function on GCP which generates a long voiceover.
  3. Again, the post content is retrieved, updated with the new audio, and a final email notification is sent.

Audio Integration on WordPress

Audio is automatically added to the post with this HTML snippet:

htmlCopyEdit

🎧 VoiceOver: {{1.title}}


Date: {{66.date}}

Article link: {{1.link}}

Post ID:{{1.id}}



This allows users to listen to the content directly on the page, making the blog accessible, usable and modern.

Security & Best Practices

Being an articulated flow, I have placed maximum attention to safety:

  • Using API keys, OAuth and tokenized access,
  • Cloud Storage Bucket with strictly controlled permissions,
  • Logging via JS for internal controls and fallback,
  • Modularity and complete traceability of each step.

Each step is verifiable, idempotent and can be isolated during debugging.

✅Conclusions

This workflow is the result of months of optimization and real-world testing.
It allows me to publish content on WordPress and, without any manual intervention:

  • archive it,
  • improve it,
  • turn it into audio,
  • and make it publicly available.

It is a solution that unites intelligent automation And attention to cybersecurity, designed for those who work with content and want to save time without giving up control.

If you want to implement something similar, or discuss automations for your stack, write to me!