How to Make a Podcast Intro With AI: Script, Voice and Music

To make a podcast intro with AI, write a script of about 35 words, turn it into a voice, generate a short instrumental theme, then join the pieces, balance their volumes and export. That's the whole job, and you can do every step in the browser.

I built a 26-second intro for a made-up gardening show this way, using MyEdit's AI tools and measuring the loudness of each layer, so you can see what each tool gives you and where you'll still have to make a call. Tested on November 2, 2026.

A podcast intro tells new listeners what show they're in and why they should stay, in as few seconds as possible. Most intros mix the same few parts, in different orders:

• A hook: a question or promise that makes someone keep listening.
• The show's name and promise: what it is and what listeners get every episode.
• A theme: a short piece of music that people learn to recognize.
• A handoff: a line or a musical ending that moves into the episode.

Here's how those parts sat in the intro I made. The music-first order is the one MyEdit's Podcast Studio supports, because it places clips one after another. I measured each part's timing and loudness on the exported MP3:

Write your intro script before you touch any audio, because its length decides how long everything else runs. In my test, a preset AI voice read about 3 words per second, so 35 words filled 11.5 seconds. That's a useful budget: about 30 to 40 words if you want the voice part to stay near 12 seconds.

A simple formula that fits that budget:

1. Hook (one sentence): ask the question your listener already has.
2. Name and promise (one sentence): say the show's name and what it covers.
3. Format (one short sentence): what every episode gives them.
4. Handoff (two or three words): a line that starts the episode.

This is the script I used for the made-up show Pocket Garden:

“What if your balcony could feed you all summer? This is Pocket Garden, the show about growing real food in small spaces. Every week: one plant, one problem, and one simple fix. Let's dig in.”

Read your draft out loud with a timer before you generate anything. If it runs past 15 seconds, cut the format sentence first: the hook and the show's name matter most.

You can make every layer of the intro with MyEdit's AI tools and put them together in its Podcast Studio, all in a browser. MyEdit is our own product, and the same workflow works with any AI voice, music generator and audio editor you already use. Each step below uses a different tool, so each one comes with its own screenshot.

Open Text to Speech, paste your script and pick a preset voice by language, gender and style. Click Generate, listen, then download the audio. On November 2, 2026, the editor showed a cost of 2 credits per 1,000 characters, so a 190-character script used 2 credits.

Use a preset voice or your own: the tool can also clone a voice from a short recording, but never clone someone else's. Our download was a mono MP3 at −21.2 LUFS, quieter than the music you'll add next.

In AI Music Generator, describe the mood, tempo and instruments, then turn on Instrumental Only so no singing competes with your voice. Ask for two tracks so you can choose: each track cost 5 credits in my test, and the tool can make up to six at once.

The shortest length you can set is 60 seconds, and my tracks came back at 58 and 59 seconds. That's far longer than an intro needs, so you'll trim it in step 5. Music generated with MyEdit can be used in personal and commercial projects, including your podcast.

A sting is a one- or two-second sound that opens the intro. In AI Sound Effect Generator, describe it in a sentence (up to 800 characters) and generate. My prompt returned three 10-second variations, so pick one and plan to keep only its first second or two.

This layer is optional. When I played the finished intro back, the sting was the part that felt out of place, so try your intro with and without it before you keep it.

Open Podcast Studio and click Add Audio, then Import From Local to bring in your voice file. For the theme and the sting, open the Music panel and click Import Music. Tracks you made in other MyEdit tools didn't appear in Podcast Studio's library during my test, so download them first and import them as files.

Each time you add a clip, Podcast Studio asks whether to place it in front of or behind the selected clip. Put the music in front of the voice, then the sting in front of the music.

Open each clip's menu (the three dots) to reach Trim, the volume slider and the Fade In and Fade Out switches. Here's what I set:

• Sting: trimmed to 2 seconds, volume 51%, fade out.
• Theme: trimmed to about 6 seconds, volume 33%, fade out.
• Voice: left at 100%.
• Theme tail: the theme copied and pasted after the voice, volume 33%, fade in and fade out.

The volume matters more than it looks. The theme came out of AI Music Generator at −11.8 LUFS and the voice at −21.2 LUFS, so at equal settings the music would have been about 9 LU louder than your host.

Click Export & Share, choose MP3 or WAV and download. My MP3 came out as a 26-second mono file at 48 kHz and 192 kbps. Keep the intro as its own file, so you can place it at the start of every episode.

Podcast intro music should sit clearly below the voice wherever the two are close together, and the whole intro should match the loudness of your episode. In my export, the theme played at −24.5 LUFS before the voice and the voice at −21.5 LUFS, so the music stayed about 3 LU quieter.

The whole intro, though, measured −22.6 LUFS, about 6.6 LU quieter than the level Apple recommends for podcasts. Podcast Studio's export has no loudness target to pick, so the level you hear is the level you set with the clip sliders.

Here's what Apple recommends, as listed on its Audio requirements page on November 2, 2026, and what each point means for your intro:

Each clip's volume slider in Podcast Studio goes up to 200%, so you can raise a quiet intro. Do it in small steps and check the result, because louder clips also mean higher peaks. I didn't re-export at higher settings, so treat the target as something to check, not a number Podcast Studio sets for you. Apple's full recommendations are on its Audio requirements page.

The workflow gets you a usable intro quickly, but it has limits you should know before you start:

• No music under the voice. Podcast Studio places clips one after another, and MyEdit's Audio Editor opens the same editor. My own preference, after listening, was a theme that fades in and out under the voice at a very low level, and that needs an editor with separate tracks.
• Levels don't match out of the box. The music was 9 LU louder than the voice, and the export didn't bring the intro up to podcast loudness.
• Themes come long. AI Music Generator's 60-second minimum means trimming every time, and a hard trim can cut a phrase mid-note, so keep the fade on.
• Files don't move between tools on their own. I had to download each layer and import it into Podcast Studio.

I rated the result 3.5 out of 5 by ear. What held it back was the sting, which sounded out of place, and the music-first order: I'd have liked the theme to fade in and out under the voice instead.

You can record your lines straight into Podcast Studio with Start Recording, then add the theme around them the same way. A recorded voice changes two things: you'll want a quiet room and a mic close to your mouth, and its loudness will differ from the AI voice's, so check the clip volumes again.

If your recording sounds dull, as if you spoke through a pillow, that's a separate problem with its own fixes. The guide on how to fix muffled audio covers the usual causes and what an AI speech enhancer can rescue.