Podcast Intro Music: How to Get One That Doesn't Sound Like Stock

Keep podcast intro music to 5 to 20 seconds, make it sound like your show within the first two seconds, and have it get out of the way the moment someone starts talking, either by ending cleanly or by ducking under the host's voice. You can license a track from a royalty-free library, commission a composer, or generate an original with AI. For a new show that wants its own sound quickly, AI is usually the practical pick: a track generated for you in minutes rather than pulled from a shared catalog, though the provider does not guarantee no one else gets a similar output, and copyright protection is weaker. Libraries are fast, but the tracks are shared and the licenses carry fine print, while a composer gives the most distinctive result at the highest cost.

What makes good podcast intro music

Most of these lengths and levels are production rules of thumb. The loudness figure is the exception: it is Apple's published recommendation, covered below.

Why most intros sound like stock

The biggest giveaway is a popular library track that thousands of other shows also use, and the fix is an exclusive or original track. A missing signature element comes close behind. Instead of a generic "upbeat corporate" groove, choose one signature sound (a specific synth, a detuned piano, a drum pattern, a vocal chop) and build around it. Then keep it short, so nobody sits through 30 seconds of ukulele before anyone speaks. That idea overlaps with brand audio; our sonic logo guide goes deeper on making a short sound memorable.

Intro structure that hands off cleanly

A reliable pattern for a 10 to 15 second intro:

1. Hook, 1 to 3 seconds. The signature sound at full level.
2. Voice-over tag or host line, 4 to 8 seconds. Music ducks under the voice.
3. Resolve, 2 to 4 seconds. Music swells briefly, then ends or fades as the episode starts.

Narrative shows often put a cold open (a gripping clip from the episode) before the theme. Interview shows usually do the reverse: short theme, then straight into the guest introduction.

Three ways to get podcast intro music

Route 1: Royalty-free libraries, and the license gotchas

"Royalty-free" only means you do not pay per play, and the license still comes with conditions. The gotchas that matter most for podcasts, from the libraries' own terms:

• Subscription coverage ends when you stop paying. Epidemic Sound says content published during an active subscription stays cleared. A podcast intro is reused on every new episode, so if you cancel, episodes you publish afterward are not covered by that promise. Budget for the subscription as long as the show runs, or replace the intro.
• Ads and sponsored content may need a higher plan. Epidemic Sound's pricing FAQ says music can be used in ad campaigns on its Pro or Enterprise plans. If you run the intro in a promo for your show, check which plan covers that.
• Creative Commons licenses vary. The Free Music Archive's license guide explains that NonCommercial (NC) licenses forbid using the music for advertising or promoting a product or service, and NoDerivatives (ND) licenses forbid syncing the track to video, which matters the moment you upload episodes to YouTube. Attribution (BY) licenses require a credit in your show notes.
• Channel safelisting. Libraries that register their music with YouTube's Content ID usually ask you to connect or safelist your channel. Forget, and a licensed track can still trigger a claim.

Libraries are a fine way to launch. The problem is that the best tracks are the most used, so your intro may play on a competitor's show too.

Route 2: Commission a composer

A composer gives you a theme nobody else has, with revisions and usually a set of variations (full theme, short version, stingers). This is the best route for a show with a budget and a long runway. Get two things in writing: whether you receive an assignment of copyright or a license, and whether the composer will register the track with a Content ID service. The second one causes surprise claims on YouTube when a composer's distributor fingerprints the track.

Pricing varies too much by composer and scope to quote a fair number. Ask for a quote that includes stems and alternate lengths.

Route 3: Generate with AI

AI music generators produce an original track from a text description. For a podcast intro, the advantage is speed and iteration: you can hear ten directions in the time it takes to email one composer. The trade-offs are less control over exact melody, variable quality between generations, and weaker copyright protection (covered in the rights section). Our roundup of the best AI music generators for YouTube intros compares tools if you are shopping around.

Podcast intro music by show type: genre matrix with ready prompts

Each prompt is written for an instrumental intro that works under speech. Paste it into an AI music generator and adjust the bracketed signature sound to make it yours.

Ready-to-use prompts:

TRUE CRIME
Dark minimalist instrumental podcast intro, 80 BPM, low repeating piano motif in a minor key,
pulsing sub bass, distant ticking clock, subtle tape hiss, restrained and tense, no vocals,
clear space for a voice to talk over after the first 3 seconds, clean ending.

BUSINESS / FINANCE
Confident modern instrumental podcast intro, 108 BPM, muted funk electric guitar riff,
tight dry drums, warm bass, bright but not cheesy, no vocals, short 2-bar hook at the start,
then a lighter groove that can sit under a voiceover, ends on a clean hit.

COMEDY
Playful upbeat instrumental podcast theme, 120 BPM, slap bass groove, short brass stab hook,
woodblock and shaker percussion, cheeky and warm, no vocals, punchy 2-second opening,
leaves room for talking, ends with a single horn button.

TECH
Precise forward-moving electronic podcast intro, 118 BPM, arpeggiated analog synth,
crisp glitchy percussion, deep round bass, optimistic, no vocals, hook in the first 2 seconds,
simpler loop underneath for voiceover, clean stop.

WELLNESS / MEDITATION
Calm spacious ambient podcast intro, 66 BPM, soft evolving pads, felt piano notes,
light wind chimes, warm and gentle, no drums, no vocals, slow fade in and gentle fade out.

INTERVIEW / CONVERSATION
Friendly relaxed instrumental podcast theme, 96 BPM, fingerpicked acoustic guitar and Rhodes piano,
brushed drums, upright bass, open and welcoming, no vocals, recognizable 4-note guitar hook at the start,
sits comfortably under speech.

HISTORY / STORYTELLING
Cinematic curious instrumental podcast intro, 88 BPM, pizzicato strings with a solo cello melody,
light timpani, sense of discovery, no vocals, not epic, melody resolves after 12 seconds.

SPORTS
High-energy instrumental sports podcast intro, 132 BPM, big stomping drums, gang claps,
distorted bass riff, stadium energy, no vocals, 2-second drum fill hook,
drops to a lighter beat for voiceover.

NEWS / DAILY BRIEFING
Urgent neutral news podcast theme, 112 BPM, ticking hi-hat and woodblock pattern,
bright synth pulse, low bass drone, serious but not dramatic, no vocals, clean stop after 10 seconds.

How to make a podcast intro on Oakgen: one brief, a whole intro kit

The goal is a set of assets that sound related: a main intro, an outro, a short stinger for segment changes, and a transition for ad breaks. Build them all from one brief.

Step 1: Write the intro brief

SHOW: [name]                      TYPE: [e.g. interview, tech]
ONE-LINE PROMISE: [what listeners get]
MOOD WORDS (3): [e.g. curious, warm, sharp]
TEMPO: [BPM]                      KEY FEEL: [major / minor]
SIGNATURE SOUND: [one instrument or texture you will reuse everywhere]
SUPPORTING INSTRUMENTS: [2 or 3]
NEVER: [vocals, specific cliché, genre to avoid]

ASSETS NEEDED
Intro theme: 10-20 sec, hook at 0:00, room for voice after 0:03
Outro: 20-30 sec, same groove, fuller, fades out
Stinger: 1-3 sec, signature sound only
Ad-break transition: 3-5 sec, hook plus a short resolve

Step 2: Generate the intro theme

In the music generator, choose Eleven Music, paste a prompt built from the brief (the matrix above is a good start), and set the duration a little longer than you need. Eleven Music accepts durations from 3 to 400 seconds, and the slider moves in 10-second steps, so you will trim anyway. Generate three or four directions before you pick one. Lyria 3 Pro is the other option; it can take reference images (your cover art, for example) as inspiration, but it does not take a duration setting.

Step 3: Build the outro, stinger and transition

Asking for "the same song, but shorter" with the same prompt gives you a related track with a different melody. AI music models do not reliably reproduce a motif across separate generations.

So use two methods:

• Cut from the master. The stinger and ad-break transition should be edited out of your chosen intro theme: the first 1 to 3 seconds of the hook, or the hook plus the final hit. This guarantees they match.
• Generate the outro. An outro can be a separate generation using the same prompt with "fuller arrangement, 30 seconds, fades out" added. It will feel like the same show without being identical, which is fine at the end of an episode.

Trimming, fading and assembly happen in an editor such as Audacity, GarageBand, Reaper or Descript. Oakgen does not include a multitrack audio editor.

Step 4: Add a voice-over intro with text-to-speech

A spoken tag makes the intro feel produced, and it is the easiest place to use text-to-speech even if you host the show yourself. Keep it to 15 to 30 words:

"From [network or city], this is [Show Name]: [one-line promise].
Here's your host, [name]."

Generate it in the AI voice generator. Pick a voice that contrasts with the host (if you are a low-voiced host, a brighter announcer voice separates the tag from the conversation), and generate a few takes at different stability settings. If you want the tag in your own voice without recording it each time, Oakgen can clone a voice from a sample of at least 10 seconds, but only your own voice or one you have explicit permission to use. For choosing a voice tool, see our guide to the best AI voice generators for podcasters.

A sound effect under the tag (a record crackle, a newsroom hum, a door) can add character. Our AI sound effects guide has prompts.

Step 5: Mix, duck and master

• Put the hook at full level, then duck the music roughly 15 to 20 dB when the voice enters. Use sidechain compression or draw volume automation.
• Fade the music out within a sentence or two of the host starting, unless you are deliberately using a bed.
• Master the full episode, intro included, to about -16 LKFS with true peak at or below -1 dBFS. More on that below.
• Export once, listen on earbuds and on a phone speaker, and check that the first word after the intro is not quieter than the music was.

What the intro kit costs on Oakgen

Numbers are from Oakgen's live pricing config (September 2026). You'll always see the credit charge before a generation starts.

Four 23-second intro directions, one 33-second outro and three voice-over takes come to roughly 200 credits, about a tenth of the 2,000 credits the Basic plan gives you each month. Current plans are on the pricing page.

Podcast loudness targets for intro music

Mixing the intro separately and pasting it onto every episode is how shows end up with a theme that blasts and a conversation that whispers. Master the full episode to one target.

We could not find a current Spotify page that publishes a podcast-specific loudness target, so we are not quoting one. In practice, mastering episodes to about -16 LKFS with -1 dBFS true peak meets Apple's recommendation and stays close to Spotify's playback level. Measure with any ITU-R BS.1770 loudness meter; most editors and free plugins include one.

YouTube intro music: what changes

If the podcast also goes on YouTube, the same theme works once you adjust for the platform:

1. Shorten it. Video viewers can skip ahead instantly. A 3 to 7 second sting before the first line, or a cold open with the theme after it, respects that. The same logic applies to channel intros in general; our YouTube intro music comparison covers standalone channel themes.
2. Check Content ID before you upload. YouTube's Content ID automatically scans uploads against audio registered by rights holders. A match can block the video, monetize it for the claimant, or track it, and the result can differ by country.
3. Know what "copyright-safe" means on YouTube. YouTube's help center says music from the YouTube Audio Library will not be claimed through Content ID, and that only Audio Library music is known to YouTube to be copyright-safe. It also says YouTube is not responsible for problems with "royalty-free" music from other libraries.

Content ID risk with library music

Library music is the most common cause of surprise claims on podcasts. The pattern: a composer uploads a track to a library and also registers it with a Content ID distributor. You license the track legitimately, upload an episode, and receive a claim anyway. Libraries that support YouTube usually fix this through channel safelisting (Epidemic Sound describes connecting your channels once you subscribe). If a claim is wrong, YouTube provides a dispute process.

An original AI-generated track carries a lower risk because it is not a registered catalog track. The risk is not zero: someone else could register similar or identical audio. Keep your generation records (the prompt, the date, the downloaded file) so you can dispute a mistaken claim.

Commercial rights: who owns AI podcast music

Can you use it commercially? Oakgen's terms say it doesn't claim ownership of your inputs or eligible outputs. Whether a given commercial use is allowed comes down to those terms, the law, the model provider's policies, and your rights in anything you put into the prompt. Oakgen doesn't guarantee that outputs are original, non-infringing or cleared for any particular use, so check the terms for your plan and model.

Can you stop others from copying it? In the US, the Copyright Office's January 2025 report on AI copyrightability concluded that AI outputs are protected only where a human author has determined sufficient expressive elements, and that providing prompts alone is not enough. Human creative arrangement or modification of the output can count. Practically: a raw AI intro may be hard to enforce against copycats. Editing, arranging, layering your own recorded elements, or having a musician build on it strengthens your position.

Can you use a real artist's style or name? Do not prompt for a specific living artist's sound or reference copyrighted songs by name. It is a fast way to get something too close to an existing work.

None of this is legal advice. For a show with sponsors or a network deal, get the music rights reviewed by a professional.

Beyond the intro

The intro is the most repeated piece of audio your show owns, so treat it like brand work. For short-form clips of your episodes, our guide to royalty-free AI music for Reels and TikTok covers the vertical-video side. If you sell ad reads or run branded spots between episodes, our radio ad script template has scripts and specs for 15, 30 and 60 second audio ads. And if you want a sung tagline rather than an instrumental, start with how to make a jingle.

Sources

• Apple Podcasts for Creators: Audio requirements (accessed 2026-09-26)
• Spotify for Artists: Loudness normalization on Spotify (accessed 2026-09-26)
• Spotify Advertising: Audio ad specs and requirements (accessed 2026-09-26)
• YouTube Help: How Content ID works (accessed 2026-09-26)
• YouTube Help: Use music and sound effects from the Audio Library (accessed 2026-09-26)
• Epidemic Sound: Pricing and licensing FAQ (accessed 2026-09-26)
• Epidemic Sound: Music for YouTube (accessed 2026-09-26)
• Free Music Archive: License guide (accessed 2026-09-26)
• US Copyright Office: Copyright Office Releases Part 2 of Artificial Intelligence Report (NewsNet 1060) (accessed 2026-09-26)
• ElevenLabs: Terms of Use, output uniqueness (accessed 2026-09-26)
• Oakgen Terms of Service, inputs and outputs section (accessed 2026-09-26)
• Oakgen model and pricing configuration for Eleven Music, Lyria 3 Pro, ElevenLabs V3 and ElevenLabs Sound Effects (checked 2026-09-26)