How to create Your Own Podcast Using AI

In today’s digital age, creating a podcast doesn’t have to be complicated, especially with the help of AI. From generating the script to creating the audio and piecing everything together, AI tools can handle it all.

In this step-by-step guide, I’ll show you exactly how you can create your own AI-generated podcast using a few tools.

Let’s get started!

What We Will Be Creating

Before we dive into the details, let me give you a sneak peek of what we’ll be creating.

Here’s an example of a podcast episode generated using AI:

“Our current trajectory, Emily, our listeners often hear about rising temperatures, but what does this really look like on a global scale? Imagine a world where seasons as we know them have blurred. In 2034, summers in many regions are not just warmer, they’re lethally hot with temperatures regularly exceeding 40°C. And it’s not just about temperature — the heat affects precipitation patterns too, prolonging droughts in areas like the Midwest, which once served as the breadbasket of the world but now struggles to produce basic crops.”

This sounds like a regular podcast, right?

Most listeners wouldn’t guess that this podcast was created entirely by AI.

In fact, I tested this on 20 to 30 people, and none of them suspected it was AI-generated.

Tools You’ll Need:

To create your AI podcast, we’re going to use three essential tools:

1. ChatGPT (for script generation)
2. Eleven Labs (for text-to-speech)
3. Descript (for putting everything together)

Step 1: Generating the Script with ChatGPT

The first step is to generate your podcast script using ChatGPT. I use a prompt designed for “persona prompting.”

If you’re unfamiliar with this, I have a complete video on it, which you can check out for more details.

Here’s the basic prompt I use:

Generate a 1-minute podcast script where three people discuss climate change, each contributing their unique perspectives.

Once the prompt is entered, ChatGPT will generate a script. This will form the foundation of your podcast.

Step 2: Creating Voices with Eleven Labs

With the script in hand, it’s time to convert the text into audio using Eleven Labs. This software is great for text-to-speech and offers multiple voice options.

Here’s how to do it:

• Voice Selection: You need to assign different voices to each character in the script. For example, if you have characters named Alex, Jordan, and Riley, you’ll need three distinct voices. Eleven Labs has a variety of voices available. Here’s how I selected mine:
  • Alex: After testing a few voices, I chose a more narration-style voice with clear articulation.
  • Jordan: I went for a calm and thoughtful voice.
  • Riley: I selected a slightly more energetic voice to match their role in the conversation.
• Voice Settings: Eleven Labs allows you to adjust the voice settings for stability, clarity, and style:
  • Stability: I keep this as high as possible for consistency.
  • Clarity: Since this is for a podcast, I prefer a noise-free, polished output.
  • Style: I usually set this at around 10-20% for accents, to avoid making the voice too robotic.

Once you’ve adjusted these settings, you can preview the audio.

When satisfied, click Generate to start creating the audio files for each character.

Step 3: Downloading and Saving Audio Files

As the audio files are generated, download them one by one. For each character, download the corresponding audio for their part of the script.

Step 4: Putting It All Together in Descript

Now that we have all the audio files, we need to stitch them together using Descript.

Here’s a quick rundown of how I do it:

Open Descript: If you’re new to Descript, it’s an excellent tool for working with audio files. You can use either the web version or the desktop version (I recommend the desktop version for faster performance).

Create a New Project: Click on New Project and select Audio Project since we are only working with audio for this podcast.

Import Audio Files: Next, import the audio files for each character. On the right-hand side of Descript, you’ll see the files being uploaded.

Speaker Assignment: Descript will prompt you to assign speaker names. Label each file with the correct character name (e.g., “Alex,” “Jordan,” “Riley”).

Transcription: Descript automatically transcribes the audio files back into text, which makes it easier to align everything with your original script.

This transcription is key because we’ll be working with the text as if it’s a document.

Step 5: Arranging the Podcast Script in Descript

Now comes the part where we arrange the podcast dialogue to match the script:

Cut and Paste: Start by cutting and pasting the text for each character’s dialogue into the correct order. For example, if Alex speaks first, copy Alex’s dialogue and paste it into the timeline.

Matching Dialogue: Continue this process for each character, matching the text and audio as per the original script.

Descript makes it easy because you can see who is speaking and rearrange their dialogue in a simple, intuitive way.

Final Touches: Once the text is in place, play the audio to ensure everything flows smoothly. You can make small adjustments to timing if needed.

Step 6: Exporting the Final Podcast

Once everything is in order, you’re ready to export the final audio file. Descript allows you to export in different formats, but MP3 is the standard for podcasts.

Simply click on Export and select your preferred settings.

Optional Features in Eleven Labs

Before we wrap up, let me tell you about some extra features in Eleven Labs that you might find useful:

Multiple Accents: You can choose from accents like English, American, British, Swedish, etc., to give your podcast a unique touch.

Voice Cloning: This feature lets you create a clone of your own voice, which can be especially useful for people who don’t have the time to record their own podcasts.

Conclusion

You now have a complete podcast generated using AI. From scriptwriting with ChatGPT to voice creation with Eleven Labs and audio editing with Descript, these tools make podcast creation easy and accessible. Try it yourself and see how AI can transform your podcast production process.

If you’d like more tips on AI podcasting, or want to explore more tools like these, feel free to check out my AutoPod AI editing tool.