Where to Find Royalty-Free Sound Effects and Music for TikTok, Reels and Shorts Ads

Where to Find Royalty-Free Sound Effects and Music for TikTok, Reels and Shorts Ads

For short-form ads, use audio that is explicitly cleared for commercial use: TikTok's Commercial Music Library for TikTok, the YouTube Audio Library for YouTube, licensed libraries such as Pixabay, Freesound (CC0), Epidemic Sound or Splice, or sounds you synthesize yourself. Trending sounds from a platform's general library are usually not licensed for brand content.

The second half of the job is fit. A good ad bed is cut to the exact ad length with hits on the frames where your visuals change.

What does "royalty-free" actually mean?

Royalty-free means you pay once (or nothing) for the right to use a sound, rather than paying a royalty per play, view or broadcast. It does not mean copyright-free, and it does not always mean free of charge.

A commercial use license covers content that promotes a business, product or service, which includes paid ads and most branded organic posts.

Read three things in every license: whether commercial use is allowed, whether attribution is required, and which platforms or placements it covers.

Can I use trending TikTok sounds in an ad?

Usually not. TikTok's Commercial Music Library user terms say that Commercial Sounds are the only sounds made available on TikTok for commercial uses. Business accounts see only the Commercial Music Library when they tap Add Sound, as TikTok's help article explains.

Each Commercial Sound lists its Usable Placements. If you plan to run the same cut on Reels or Shorts, check that the track is cleared there too, or pick a source that licenses across platforms.

Which royalty-free sources work for short-form ads?

Here is a quick map of common sources and what to check on each.

• TikTok Commercial Music Library: free for businesses on TikTok. Check Usable Placements and region.
• YouTube Audio Library: free music and sound effects in YouTube Studio. YouTube's help page says tracks downloaded there won't be claimed through Content ID, and Creative Commons tracks need credit in the description. It is built for YouTube videos, including Shorts.
• Pixabay music and sound effects: free, no attribution required under its license summary, with limits such as no reselling the files on their own.
• Freesound: a huge community library of sound effects. Licenses vary per sound (CC0, Attribution, Attribution-NonCommercial), so filter for CC0 or Attribution for ads and skip NonCommercial. The Freesound FAQ explains each one.
• Splice: a subscription sample library. Splice's licensing FAQ covers using samples in your own productions, which suits editors who build their own beds from loops and one-shots.
• Epidemic Sound: a subscription library whose commercial plans cover ads on TikTok and other platforms.
• Synthesized sounds: sounds rendered from your own code have no third-party samples inside. Opus Sounds Directory collects free ad beds by length, risers, drops and logo stings, each with its prompt and Python code, so you can re-render a sound to your own timing.

For example, a 20-second bed with a hook at the 10-second mark, sized for a short product ad:

What lengths and cue points do short-form ads need?

Cut beds to the placement length, not "roughly" to it. Common lengths are 6 s (bumper), 15 s, 30 s and 60 s.

At 30 fps and 48 kHz, those lengths are:

Length	Frames at 30 fps	Samples at 48 kHz
6 s	180	288,000
15 s	450	720,000
30 s	900	1,440,000
60 s	1,800	2,880,000

Put a hit on frame 0 so the first second grabs attention, place a change (drop, switch, reveal) where the hook pays off, and end on a real final beat. A music bed is background music cut to sit under picture and voice for a fixed length.

What sound effects does a short-form ad usually need?

Most ads reuse the same handful of sounds:

• Whooshes on cuts and swipes.
• Risers that build tension into a reveal.
• Impacts or drops on the reveal itself.
• UI sounds (taps, pops, success chimes) for app demos.
• A logo sting for the end card.

Keep them short and on the frame. A whoosh that starts two frames early feels sloppy even if viewers can't say why.

How do I generate a bed that fits the ad exactly?

If no library track fits, write a numeric prompt for a code-capable LLM. This one makes a 15-second upbeat bed, using the same template as the 15s upbeat retail pulse entry:

Make a 15 s upbeat ad bed for a vertical video ad at 30 fps.
Output: one WAV, 48 kHz / 16-bit / stereo, exactly 720000 samples.
Python + numpy/scipy only. Synthesise everything (no samples, no downloads). Fixed random seed.
Timing: 128 BPM. Cues: frame 0 = hit; frame 210 = drop into full groove; frame 420 = tag/end card;
final beat ends cleanly and fades to exactly 0 by sample 720000.
Sound: bright synth plucks, four-on-the-floor kick, claps on 2 and 4, warm bass, key E major.
Energetic but not harsh: no sustained energy above 10 kHz.
Master: -14 LUFS integrated, true peak <= -1 dBTP.
Verify length, loudness and peak with ffmpeg ebur128; confirm energy changes at frames 210 and 420;
render a spectrogram. Fix and re-render until every check passes, then report the numbers.

And a 6-second bumper with a reverse swell into a hit, close to the 6s bumper sting entry:

Make a 6 s ad bumper for video at 30 fps.
Output: one WAV, 48 kHz / 16-bit / stereo, exactly 288000 samples.
Python + numpy/scipy only. Synthesise everything. Fixed random seed.
Sound: 0.5 s reverse-cymbal-style noise swell into a punchy synth chord + tight kick at frame 15,
short groove, resolve on a sustained chord by frame 150, silence-safe tail to the end.
Master: -14 LUFS integrated, true peak <= -1 dBTP.
Verify length, loudness and peak with ffmpeg ebur128 and report the numbers.

Note the 128 BPM grid does not divide evenly into 30 fps frames (one beat is 14.06 frames). That's fine for music, but put hard cues on whole frames.

How loud should an ad mix be?

Aim for about -14 LUFS integrated and a true peak no higher than -1 dBTP for the full mix, then listen on a phone. Platforms normalize and process audio in different ways and don't all publish a single number, so the goal is a mix that is clear and doesn't clip.

Measure with ffmpeg's ebur128 filter. If voiceover is present, keep the bed well underneath it.