Not everyone wants to read your blog. Some people want to listen to it on a walk, during a commute, or while doing dishes. Turning written content into audio is one of the highest-leverage things you can do for reach, but recording, editing, and publishing narration for every single post is its own part-time job.
So I automated the entire pipeline. Now, the moment I publish a new blog post, it's rewritten for the ear, narrated, and embedded back into the article as a playable audio player, with zero manual work.
Here's exactly how it works
and how to set up the same thing for your own blog. If you'd rather skip ahead, Grab the workflow from here
This runs on n8n, using Google Gemini to rewrite the article for spoken word and Smallest.ai to turn that script into natural-sounding audio.
The rewrite step matters more than people expect. A blog post written for the eye is full of things that sound terrible read aloud — bullet points, headings, stray URLs, abbreviations. So instead of just piping raw text into a TTS engine, this workflow rewrites it into something that actually sounds like a person talking, then hands it to Smallest.ai, which is what makes the final audio sound like a real narration instead of a robotic reading of a web page.
How the workflow actually works
Watch the RSS feed the workflow polls your blog's RSS feed on a schedule. The moment a new post shows up, it kicks off.
Strip it down to clean text, a code step strips out all the HTML, formatting, and leftover markup, leaving just the raw article text to work with.
Rewrite it for the ear, a Google Gemini agent rewrites the post specifically for narration: bullet points become flowing sentences, numbers and abbreviations are spelled out, stray captions and credits are removed, and it adds a natural spoken intro and outro. The result reads like a script, not an article.
Synthesize the narration, that script is sent to Smallest.ai, which generates a clean MP3 narration of the piece.
Host and embed the audio, the audio file is uploaded to Google Drive, made publicly accessible, and an audio player is automatically embedded at the top of the original WordPress post, so readers can either read or listen.
New post published → rewritten, narrated, hosted, and embedded, all before you've finished your coffee.
Why this setup is worth copying
It multiplies your content without multiplying your effort. Every post you already write becomes an audio piece for free.
It sounds like narration, not a text reader. The rewrite step is what separates a listenable article from a robotic wall of text being read aloud.
It's a real accessibility win. Not everyone can or wants to read, this gives every post an audio alternative by default.
Setting it up yourself
Point the RSS Feed Trigger at your blog's feed URL.
Authenticate Google Gemini inside the LangChain chat model node.
Add your Smallest.ai API credentials to the speech synthesis node.
Connect your Google Drive and WordPress accounts for hosting and embedding.
Want a different narration tone, a custom intro/outro, or a shorter target length? It's a single prompt edit inside the AI Agent - Rewrite for Audio node.
Final thought
Repurposing content into audio is exactly the kind of task that's valuable but tedious enough that most people never actually do it consistently. Pairing n8n's orchestration with Smallest.ai's speech synthesis turned it into something that just happens in the background, every time you hit publish.
If you build on top of this, I'd love to see what you add to it.
Top comments (0)