What audio format does TikTok support?

TikTok only accepts audio embedded inside a video file — MP4 or MOV containers with AAC-LC audio. There is no path for uploading standalone MP3, WAV, or FLAC files through the consumer app or Creator Center. The one exception is music distribution: a release submitted through DistroKid, TuneCore, or a similar service ingests as 16-bit/44.1 kHz WAV or FLAC, which TikTok transcodes server-side into the public sound library.

Does TikTok compress audio when I upload?

Yes, every upload is transcoded into at least three quality tiers — high (around 128 kbps AAC stereo), medium, and low — so the player can adapt to network conditions. The original file is not preserved. To minimize damage, upload AAC at 256–320 kbps so TikTok's encoder works from a near-transparent source rather than re-encoding lossy MP3 or low-bitrate AAC. Lossy-to-lossy hops compound quality loss far more than a single transcode from a high-quality source.

Should I upload 44.1 kHz or 48 kHz audio to TikTok?

Either works and TikTok will not penalize you for choosing one over the other. 44.1 kHz is the music industry standard and is fine for purely musical content. 48 kHz is the video standard and is what every camera, NLE, and broadcast pipeline uses by default — so if your audio came out of Premiere, Final Cut, or DaVinci Resolve, it is already 48 kHz and there is no benefit to resampling down. Pick whichever rate matches your source and avoid an unnecessary sample-rate conversion.

How long can a TikTok sound be?

Videos can be up to 10 minutes, but the clipped sound that other users can adopt is capped at 60 seconds for original audio uploaded with a video, and 30 seconds for tracks delivered through music distribution. TikTok's algorithm surfaces sounds that get reused, so the most hooky 15-second segment of your audio should land near the start of the video — that is what TikTok pre-clips as the default 'Use this sound' snippet.

What loudness target should I master for TikTok?

Aim for -14 LUFS integrated with true peaks below -1 dBTP. TikTok normalizes playback to roughly -14 LUFS like Spotify and YouTube, so masters louder than that get turned down (which makes over-compressed mixes sound dull) and quieter masters get turned up (which raises the noise floor). Use Youlean Loudness Meter, iZotope Insight, or Logic's built-in loudness meter to measure. Voiceover content tolerates a slightly louder target — around -12 LUFS short-term — to stay above ambient phone-speaker noise.

Best Audio Format for TikTok: Specs and Upload Tips

TikTok does not accept standalone audio files. Every piece of audio that ends up on the platform either travels inside a video upload (MP4 or MOV with embedded audio) or arrives through music distribution into TikTok's commercial sound library. Both routes end with AAC playback, but they are governed by completely different rules around quality, length, and licensing.

How TikTok's Audio Pipeline Works

When you upload a video, TikTok's servers transcode the file at least three times: a high-bitrate version for fast connections and modern devices, a medium version for typical mobile, and a low version for poor networks. The output is always AAC-LC at 44.1 kHz or 48 kHz, typically capped around 128 kbps stereo for the high-quality variant. Watermarked downloads come out lower, around 64 kbps mono.

Two practical consequences. First, your upload is going to be re-encoded no matter what you do, so the goal is to give the encoder the cleanest possible source. Second, transcoding from a lossy source loses more quality than transcoding from a near-lossless source — a 96 kbps MP3 fed into TikTok's pipeline ends up worse than a 320 kbps AAC of the same recording.

Recommended Upload Specs

Container: MP4 (preferred) or MOV
Audio codec: AAC-LC, 256–320 kbps
Sample rate: 44.1 kHz or 48 kHz
Bit depth: 16-bit (24-bit gets dithered down anyway)
Channels: Stereo for music, mono for talking-head voiceover
Loudness target: -14 LUFS integrated, true peak below -1 dBTP

TikTok publishes few hard specs publicly. The numbers above match what survives the transcoder cleanly based on creator testing and what the iOS app produces from its own editor. Mobile recording from inside the TikTok app captures at 48 kHz AAC, which is why content shot in-app sounds consistent.

Sound Length Rules

TikTok distinguishes "videos" (up to 10 minutes) from "sounds" (the clippable music or audio that drives the For You page).

Original sounds are pulled from your video upload automatically and clipped to the first 60 seconds when other users adopt them as a sound.
Library sounds delivered through commercial distribution are typically 30 seconds in the public clip, even if the master is 3:30.
Voiceover and CapCut pre-baked audio rides along with your video and is not separately distributable as a sound.

If virality through sound adoption is the goal, the most replayable hook should land in the first 15 seconds. TikTok's own creator tools surface that segment as the "Use this sound" snippet by default. If your master starts with a long intro, cut the audio so the hook hits within the first second before muxing it into the video — TikTok will not move the start point for you.

Avoid Lossy-to-Lossy Re-Encoding

If your video editor outputs MP3 audio in the MP4 wrapper (some older versions of Movavi, Filmora, and certain mobile editors do), TikTok decodes the MP3, then re-encodes to AAC. Each lossy hop loses information. Exporting with AAC audio in the first place keeps the chain at one re-encode.

Premiere Pro 2024+: Export > H.264 > Audio tab > AAC, 320 kbps, 48 kHz
Final Cut Pro: Share > Export File > Settings > Audio: AAC 256 kbps
DaVinci Resolve: Deliver > Format MP4 > Audio Codec AAC > Bitrate 320 kbps
CapCut desktop: Export > Audio quality "Recommended" or higher
iMovie: Share > File > Quality "High" (forces AAC at 256 kbps)

For audio-only sources you want to slap onto a video before upload, convert to AAC first using MP3 to AAC or WAV to AAC, then mux into the video.

Distributing Music to TikTok's Sound Library

Releasing a song through DistroKid, TuneCore, CD Baby, UnitedMasters, or Amuse pushes the master to TikTok's commercial library, where any user can clip it into their video. The submitted file should be 16-bit/44.1 kHz WAV or FLAC — not MP3. Distributors transcode to whatever format TikTok ingests internally, but starting from lossless gives them flexibility for re-delivery to other platforms.

Licensing implications:

Library music carries a commercial license — you keep ownership but TikTok can serve clips to any user.
Original sounds uploaded through your own video are licensed only for re-use within TikTok itself, not for distribution.
Cover songs require mechanical licenses (Easy Song Licensing, HFA) before they hit the library.
Sample-based tracks must clear samples or the distributor will reject ingestion.

Loudness Normalization

TikTok normalizes playback to roughly -14 LUFS integrated, in line with Spotify, YouTube, and Apple Music. Loud masters get turned down without re-encoding, so over-compressing the master only hurts dynamics with no perceived loudness gain. Quiet masters get turned up, which raises the noise floor — particularly painful for spoken-word content recorded in noisy rooms.

For mixing music or voiceover specifically for TikTok, target -14 LUFS integrated and -1 dBTP true peak. iZotope Insight, Youlean Loudness Meter, and the loudness panel in Logic Pro all measure this directly.

Common TikTok Audio Mistakes

Things that cost creators reach and quality:

Uploading low-bitrate MP3 in MP4. Some older mobile editors export MP4 with 96 kbps MP3 audio. TikTok's encoder transcodes that to AAC, doubling the lossy damage. Force AAC export at 256+ kbps from your editor.
Mismatched sample rate. Some camera apps record audio at 44.1 kHz but the camera video is 48 kHz. Editors sometimes force-conform to one and resample the other audibly. Conform on a single rate (48 kHz preferred for video).
Clipped peaks. Mastering at 0 dBFS sounds loud in a quiet room but causes audible clipping after TikTok's normalization re-encodes. Target -1 dBTP.
Stereo voiceover. Solo voice content gains nothing from stereo encoding. Convert to mono before upload to halve the audio bitrate budget for the encoder, leaving more bits for clarity.
Heavy noise reduction. Aggressive noise reduction (Krisp, RX Voice De-noise on max) introduces artifacts that survive TikTok's re-encode. Light de-noise plus a high-pass filter at 80 Hz is usually cleaner.

For broader context on bitrate decisions per platform, see the audio bitrate guide by use case and the best audio format for Instagram for the cross-platform comparison. For the conversion mechanics see container vs codec explained, which covers why MP4 with AAC inside survives TikTok's pipeline cleanly while MP4 with MP3 inside compounds damage on every transcode.