Posted by Apple Podcasts ¦ Amazon Music ¦ Spotify
Hi, and welcome to Try AI for Growth, a podcast out of Make Space for Growth.
Here, I share short and sometimes surprising stories about how I’ve used technology to tackle everyday challenges—at home, at work, and in business. I’m your host, Sara Vicente Barreto, and today I want to tell you about a small experiment that helped me solve a problem many of us face every week: finding the time to sit down and write our thoughts.
My blog article last week started in a very different place than usual. Not at my desk. Not with a blank page. It started in the car.
The Problem
Like many people who write regularly, I often have ideas for blog posts… but not always when I am sitting down neatly with pen and paper or my laptop. Sometimes the thinking happens while you’re driving, walking, or doing something else entirely. And if you don’t capture it quickly, the idea disappears or it just doesn’t come across in the same way as you originally thought of it.
So this week I tried something simple: I recorded a voice note on my phone while driving, talking through the idea for the blog post exactly as it came to me. With voice notes, I always worry about getting them exactly as a blog from the get-go, but this time, I just let it flow and even expressed some doubts on parts to include along the way. No structure. No editing. Just thoughts.
That left me with a messy voice recording—not exactly a blog post. But it was a first step I had not taken before. And I knew exactly what I was going to do with it.
The Experiment
AI was no doubt my partner for this new “writing” experiment. The first step was to upload the audio into Descript, which is a tool that can automatically turn speech into a transcript. And the quality was very good.
I use Descript on my computer, so there was an intermediate step of sending my voice note to my email so I could get the audio file. When uploaded to Descript, within minutes, I had a written version of everything I had said in the car.
Now, if you’ve ever looked at a raw transcript, you’ll know it’s not pretty. There are half-sentences. Repeated words. Thoughts that start in one direction and end somewhere else entirely. And quite a few starts and stops.
It was time to call my editing partner – ChatGPT.
“This is a transcript of a voice note where I’m brainstorming a blog post. Please clean it up into a structured draft with no duplications and hesitations, and keeping my voice. Do not change any words unless there are grammatical errors. Do not add any themes or ideas as I will use this as my first draft. ”
And suddenly, what had started as a messy stream of thoughts became a first draft on my computer. Not perfect. But very usable. Ready to be worked on.
Why This Works
I really benefited from this new workflow. Often, I am not able to set aside time for writing, or I get unexpected meetings. This time, I was driving to a breakfast meeting, and I wanted to get started as early as possible. And so I did. This process was a quick way to solve 2 problems:
- Capturing ideas quickly: Speaking is often faster and more natural than writing. When you’re talking through an idea, you don’t get stuck on wording; you just explain what you’re thinking. As soon as I accepted, I was not “writing” the final blog into the recorder, but rather just gathering a set of ideas that I would work on later; it all became much more fluid.
- Turning raw thoughts into text.: That’s where AI is extremely helpful. A tool like Descript converts the voice into a transcript. Afterwards, ChatGPT can be very good at taking messy material and cleaning it up into something coherent, with the very clear instructions not to add anything or change any content. I wanted a pure clean-up.
The combination worked smoothly: Voice, Transcript, Draft.
Once I had the cleaned-up version, I still needed to do the human part. I moved my draft into a structured blog post. I added sections, completed content, and worked on a basis that was entirely mine. Instead of starting from a blank page, I started from something that was already 80% of the way there. And that made a big difference.
Lessons from the Experiment
This small experiment reminded me of a few simple principles when using AI, and the versatility that different models bring to the workflow.
- Capture ideas in the easiest format
I am a writer at heart and that is still my favourite way of starting. But I do spend a lot of time driving (courtesy of my own life flexibility). And I was often getting frustrated that I had ideas and could not find a good way to get them going while in the car. Recording a quick voice note allowed me to get going, stay on my topic and then later address the deeper parts.
- Be flexible about how each tool can be used
I used ChatGPT sometimes for voice, but the latest versions don’t seem to give me a transcript anymore, and you can’t upload videos. So I remembered Descript, which is usually used for audio or video editing through a transcript, but the free version allowed me to get the transcript even without taking the next step to work on the video.
- Combine tools
After getting my transcript, I knew it would still take me a while to clean up all my hesitations and mumbles. Gen AI is excellent at transforming rough input into structured drafts. That was a key step in saving me time to be ready to start working on my content in depth.
Needless to say, keeping the human in the loop is a key step in all this. My transcript was still a bunch of thoughts that I wanted to come from me and how I think of things. So I did not ask ChatGPT to turn it into a blog. However, once I was done, I used my ChatGPT Agent to do an editorial review and prepare my socials. But only after I decided what I wanted to say. AI can give you a great hand, but the final voice, examples, and judgment still need to come from you.
Why I Like This Workflow
What I like most about this process is how low-friction it is. You can capture an idea anywhere. In the car. On a walk. Between meetings. And by combining tools like Descript and ChatGPT, that raw idea can quickly become something structured enough to build on. For people who write regularly and are often on the go, this workflow can be a lifesaver.
This week’s blog started as a voice note in the car, turned into a transcript with Descript, and then became a clean draft with ChatGPT. It’s a simple workflow, but it’s one that makes it much easier to capture ideas when they happen—and turn them into something useful later.
So if you often find yourself thinking “I had a great idea earlier, but now I’ve forgotten it,” try this: record it, transcribe it, clean it up. That should get you going. Sometimes the best ideas don’t start at a desk.
That’s it for today’s experiment. Thanks for listening to Try AI for Growth. If you’ve found creative ways to use AI in your daily work or life, I’d love to hear about them.
Until next time — keep experimenting and keep having fun.
Photo by Mark Twain on Pexels