When I wrote my long read about Socotra, I immediately felt that this was more than just a text story. Socotra is not a destination you can describe superficially; it is an island you have to feel, smell and hear. The dragon’s blood trees, the silence, the encounters. But how do you convey that to an audience you don’t know, who may never have heard of Socotra?

I decided to make my story audible. Not as an audiobook or a traditional podcast, but as something new: an AI-generated episode, in my own voice, professionally edited and distributed internationally via Spotify.

From curiosity to experiment

My first thought was, to be honest: is this really possible? Making a podcast without a studio, without a voice-over, and still sounding professional? I decided to just give it a try. I had already written my long read in English, partly because I wanted to share the story internationally, partly because the theme of sustainable tourism is universal.

The reason I chose a solo episode first is because I’m a fan of Gary Arndt’s podcast. He records everything solo, which creates an intimate, direct atmosphere. That fit perfectly with my story about Socotra.

After trying out and comparing various AI tools, I finally chose JellyPod.AI. What appealed to me was its simplicity. You don’t need a studio or a microphone, just a good story. And it works surprisingly intuitively. My process looked something like this:

1. Provide input
   JellyPod asked for a text, URL or outline. I used my long read as a basis.
2. Clone voice
   I read a few sentences. Based on that, the tool created a voice model of my own voice.
3. Select voice
   I chose my uploaded voice as the voice-over for the episode.
4. First outline
   Within a minute, JellyPod generated a proposal for the structure of the episode.
5. Generate script
   Once the outline was approved, a complete script was automatically generated.

Of course, that script wasn’t perfect. Some sentences sounded too journalistic, too much like written text. But the great thing about JellyPod is that you can edit everything and listen to it right away. You can immediately hear how your voice pronounces the text, where the emphasis lies and whether the rhythm and tone are right. It took some getting used to, hearing myself without being myself, but it worked. And it sounded surprisingly good for a first attempt.

I spent so long fine-tuning that first episode that I ran out of credits. Fortunately, a few weeks later, a new feature was added that allows you to edit parts of the script. That costs a lot less credits and makes the process much more efficient.

The second episode: an AI conversation

For episode two, I wanted to try something new: an interview. JellyPod also has an interview mode. You can add a second voice from a guest by uploading a short audio recording. Or you can choose a voice from the library.

I simulated a conversation with Rafa from SocotraTours. He gave permission to clone his voice. The conversation is based on his spoken WhatsApp messages, which he sent me in response to the questions I had prepared for my long read. JellyPod took care of the editing and even the intonation differences. Rafa’s voice sounds exactly like I heard him in real life. Bizarre. He loved it when I shared it with him. The result? A fluent conversation, as if it had taken place live. Within fifteen minutes. No travel time, no studio, no editing stress.

Why this works, even for destination marketing

I believe that these tools are not only valuable for creators like me, but also for city marketers, policymakers and tourism organisations. Why?

• You don’t need to be a technician or podcaster
• You can create content quickly and in multiple languages
• You have complete control over tone, content and pace
• You create stories that fit how people consume content today: on the go, visually and auditorily

Of course, some people will say that your voice isn’t quite right, or that it doesn’t sound spontaneous. But technology is developing rapidly. In no time at all, it will be almost indistinguishable from the real thing. And that opens the door for many more people to make their stories heard.

That’s exactly what destination marketing is all about: telling stories that touch and connect. Not just about a place, but about the people who live, work or travel there. Imagine letting residents tell their stories. Or turning policy plans into inspiring audio stories. Or strengthening campaigns with Spotify episodes. It’s all possible. We are at the beginning of a new way of storytelling. One in which technology is not a replacement, but rather an enhancement of stories.

What I learned from making the AI podcast

My most important insight? A good AI podcast doesn’t start with technology, but with storytelling. The tools are there, but you remain the storyteller. And perhaps even more importantly, I felt how satisfying it is to hear your own voice in a story you’ve written yourself. Not because it’s perfect, but because it’s truly yours.

The first episode took me hours. The second took an hour. And now I see dozens of possibilities for new formats, collaborations and stories. It takes time to invest in this and make the tool your own. But when you succeed and see what’s possible, you become more and more enthusiastic.

Listen to the episodes

Episode 1 – Socotra through my eyes
In this episode, I take you to the magical island of Socotra. A story about wonder, silence and the search for balance between travelling and protecting. Created with my own AI voice.

Episode 2 – Socotra a conversation with local travel agent Rafa
A virtual conversation with Rafa from SocotraTours, whose voice I cloned with his permission. We talk about what makes Socotra unique and how technology can make local voices heard.