Your intro is the first thing every listener hears, in every episode. A tight, well-produced intro makes even a home-recorded show sound professional - and with AI you can create the announcer voice, theme music and transitions yourself in an afternoon.
What makes a good podcast intro
- Short: 10 to 20 seconds. Listeners came for the episode, not the intro.
- Clear: the show name and the promise of the show in one sentence.
- Recognizable: the same music and voice every episode, so the intro becomes your audio logo.
- Well mixed: voice on top, music underneath, consistent loudness with the rest of the episode.
Step 1: Write a 15-second script
Around 30 to 40 spoken words fit into 15 seconds. A simple structure works for almost every show:
[Show name]. The podcast where [who it is for] learn [what they get].
Every week, [host] talks to [guests or topic].
Let's get into it.Example: "Build Notes. The podcast where small business owners learn how real companies grow. Every week, Anna talks to founders about what worked - and what didn't. Let's get into it."
Write out names and numbers exactly as they should be spoken, and read the script aloud once to check the rhythm.
Step 2: Generate the announcer voice
Open Text to Speech and try two or three voices with your script. For a classic "announcer" feel, pick a confident voice and lower Stability slightly for more energy; for a calm, conversational show, keep Stability in the middle and Speed at 1.0x or a touch slower. Download WAV for mixing.
Want a signature voice nobody else has? Create one from a description in Voice Design - for example "a warm, confident male voice in his thirties, neutral American accent, friendly radio host energy". Or use your own voice with a voice clone.
Step 3: Generate the theme music
In Music, turn on Instrumental and describe the identity of your show. Set the duration to around 20-30 seconds so you have room to edit.
- Business or tech: "modern upbeat electronic, clean synth plucks, punchy drums, confident and optimistic, 118 BPM"
- True crime: "dark cinematic ambient, low piano notes, tense pulsing synth, slow build"
- Comedy or lifestyle: "bright funky groove, slap bass, horns, playful and warm, 105 BPM"
Generate a few versions and pick the one with the strongest first two seconds - that opening is what listeners will remember. More tips in how to make a song with AI.
Step 4: Add a transition sting
A short sound between the intro and the conversation - a whoosh, a riser or a soft impact - signals "the episode starts now". Create one in Sound Effects: "short airy whoosh rising into a soft cinematic hit, clean, 2 seconds". The same sting can separate segments later in the episode.
Step 5: Mix it in your editor
- Place the music first and let it play alone for one to two seconds.
- Bring in the voice and lower the music under it (ducking) so the words are easy to understand.
- Let the music rise again after the last word, then end on the sting or a natural ending of the track.
- Fade out the music tail and match the overall loudness to your episode. A common target for podcasts is around -16 LUFS for stereo.
Tip: Export the finished intro as one file and reuse it. Consistency is what turns an intro into a brand.
Make an outro too
Reuse the theme with a different script: thank the listener, give one clear call to action (follow, review, visit your website) and say when the next episode comes out. A shorter, calmer cut of the same music ties both ends of the episode together.
FAQ
How long should a podcast intro be?
Between 10 and 20 seconds works for most shows. If your intro is longer than 30 seconds, listeners start skipping it.
Can I use AI music in my podcast intro?
Yes. Generate an instrumental track in Pynokio Music and use it under the Terms of your account. Because the track is generated for you, you do not need to search stock libraries.
Should I use my own voice or an AI voice?
Both work. An AI announcer voice gives a polished, radio-style feel; your own cloned voice keeps the intro personal and consistent with the episode.
Do I need to tell listeners the voice is AI?
If the voice could be mistaken for a real person who did not record it, yes - disclose it. It is good practice either way, for example in your show notes.
Give your show a sound of its own. Voice, music and stings in one place.
