iTunes, YouTube, Spotify and others – this is how your song should be mastered.
Spotify, Apple Music, YouTube, Amazon, SoundCloud and TIDAL – current status September 2026
Streaming mastering often confuses three things: the file you upload to the music distributor, the technical delivery to the platform, and the loudness normalization during playback. This guide separates these levels and shows what information the services themselves currently publish.
Why streaming services normalize loudness
As loudness war This refers to the decades-long competition for the loudest possible productions. In a brief, unbalanced comparison, a louder track often sounds more immediate. To achieve this, masters have been increasingly compressed and limited – often at the expense of transients, dynamics, and long-term listening comfort.
Louder is not automatically better: Only a level-balanced comparison fairly shows sound balance, dynamics and distortion.
Streaming services address large differences in loudness with playback normalization. This involves measuring the inherent loudness of a track and adjusting the playback level depending on the service, app, settings, and device. A very loud master track doesn't automatically gain a loudness advantage as a result, but it can still sound denser or more aggressive.
Normalization is not a subsequent improvement of the master. It primarily changes the level. Therefore, the sonic decision should be made in the mastering to be based on level-adjusted references – not on a single LUFS target value.
Loudness wins every contest except the audiophile contest, unless anyone cares how it sounds, of course. -Richard Dodd (Recording/Mastering Engineer)
Loudness normalization and upload specifications
The following overview distinguishes between publicly documented playback normalization and technical deliveryValues marked with "no fixed public target value" are intentionally labelled as follows: Apple, Amazon and YouTube do not currently publish a generally binding LUFS mastering value for all music uploads.
Important: These are not universal mastering targets. Settings, end devices, and platform clients can modify or disable normalization. You usually submit the upload file to the distributor; their specifications take precedence.
| Platform | Playback normalization | True Peak indicator | Verified delivery / playback |
|---|---|---|---|
| Spotify | Normal: -14 LUFS According to: -11 LUFS Quiet: -19 LUFS | -1 dBTP; for masters louder than -14 LUFS, Spotify recommends below -2 dBTP. | Distributor uploads are usually lossless; Spotify also offers lossless playback. |
| Apple Music | Sound Check adjusts loudness; Apple does not specify a fixed, general LUFS mastering value. | No general public music target value | 16/24 bits; 44,1 to 192 kHz; WAV, FLAC, ALAC, among others |
| Amazon Music | No currently verified, generally binding public LUFS mastering value | Follow the distributor's instructions. | Upload and catalog delivery via distributor |
| YouTube / YouTube Music | There is no universally binding public LUFS upload value; "Content loudness" is a playback indicator. | No general public mastering requirement | For partner deliveries: lossless, 48 kHz / 24 bit recommended; normal channels require a video container. |
| SoundCloud | -14 LUFS | Below -1 dBTP; for masters louder than -14 LUFS below -2 dBTP | Lossless upload from 16 bit / 44,1 kHz; HQ playback as 256 kbps AAC |
| TIDAL | No generally published fixed mastering target value | Follow the distributor's instructions. | Stereo as FLAC, HiRes up to 24 bit / 192 kHz; immersive as Dolby Atmos |
LUFS figures here describe playback normalization, not automatically the permissible upload file size or the ideal mastering value. Platforms can change their procedures at any time.
Music distributors: Prices and delivery compared
Your audio file is usually uploaded to the distributor first. Because rates, commissions, and accepted formats vary, we compare the most important providers on our site. Music distributors compared in 2026.
The standardized exchange of recordings, cover data, and metadata between distributors and platforms often takes place via DDEXTherefore, distributor specifications and the platform's subsequent playback formats may differ.
Why up to 2 dBTP true-peak headroom makes sense
Lossy encoding in AAC or MP3 can introduce additional peaks between samples. Sufficient true-peak headroom reduces the risk of these inter-sample peaks exceeding 0 dBFS and causing distortion during conversion or playback.
Spotify recommends -1 dBTP as an upper limit. If a master is louder than -14 LUFS integrated, Spotify recommends lowering the volume due to potential encoding distortion. below -2 dBTPThis is a technical safety margin, not a requirement to bring every master down to -14 LUFS.
You can find more background information in the article about inter-sample peaks.
Still have questions about the optimal sound for streaming platforms?
We will advise you on a technically clean and sonically competitive master for the relevant streaming services.
You can reach us by phone Monday to Friday from 9 am to 20 pm.
What does LUFS mean and how is it measured?
LUFS LU stands for "Loudness Units relative to Full Scale". This measure represents the perceived loudness of a digital audio signal according to ITU-R BS.1770. A change of 1 LU corresponds to a level change of 1 dB.
EBU R 128 uses the same measurement basis for broadcast programs and recommends -23 LUFS. This broadcast standard is not a streaming mastering targetFor music, the three measurement periods and the True peak value relevant:
- Momentary Loudness: Rolling measurement over 400 milliseconds.
- Short-term Loudness: Rolling measurement over 3 seconds.
- Integrated Loudness: Average loudness over the selected measurement period, with gating.
Gating prevents long periods of silence or extremely quiet passages from disproportionately lowering the integrated value. Measurements are taken in overlapping 400 ms blocks. Blocks below -70 LKFS are excluded; a preliminary value is calculated from the remaining data. Only blocks above the absolute threshold and above the relative gate of -10 LU are included in the final result.
The previously mentioned 10-to-95 percent rule is not part of the Integrated Loudness calculation according to ITU-R BS.1770. Percentiles are used, among other things, for the loudness range, but not as an upper 95 percent limit for Integrated Loudness.
More information about LUFS
What does dBTP mean – and what is True Peak?
dBTP μ stands for decibel true peak. Unlike a simple sample-peak meter, true peak measurement also estimates peaks that can occur during reconstruction between two samples. This allows for a better assessment of the risk of clipping during digital-to-analog conversion and codec conversion. EBU R 128 uses -1 dBTP as the maximum true peak recommendation for broadcast programs; streaming platforms may set their own guidelines.
Inter-sample peaks and true peak
Inter-sample peaks occur when the continuous audio signal is reconstructed from digital samples during playback. Although no single sample value exceeds 0 dBFS, the reconstructed curve may exceed this value. Lossy codecs can further amplify such peaks.
A true peak meter uses oversampling to estimate these intermediate values. A positive value, such as +0,4 dBTP, indicates that the digital maximum output level has been exceeded; a negative value describes the remaining distance.
More information about True Peak
How do streaming services analyze audio and determine the playback level?
Streaming services analyze the integrated loudness and true peak using methods based on ITU-R BS.1770. When normalization is enabled, the playback level is reduced or – if headroom is available – increased.
Spotify does not use a limiter in the "Normal" and "Quiet" settings. In "Loud" mode, a limiter with a -1 dB sample peak is used. Album and track playback may differ, and not every client normalizes: Spotify cites the web player and some third-party devices as exceptions.
More information about Spotify normalization
Streaming master for all platforms
Our analog streaming master is sonically tailored to your material and checked for loudness, true peak and codec security.
How can perceived loudness be meaningfully increased?
Compression and limiting can bring quiet and loud signal components closer together. However, more processing doesn't automatically mean more effect: Attack times that are too short, constant clipping, or excessive limiting can weaken transients, depth, and groove.
Frequency balance, arrangement, and harmonic saturation also influence perceived density. Controlled application can make vocals, bass, or midrange sound more present without simply increasing the peak level. Crucially, the processing must be level-matched to reference tracks and considered in relation to whether it serves the song.
Classifying saturation and analogous summing
Harmonic saturation can make signals sound denser and more present. It is not a blanket proof of quality and does not replace arrangement or mix balance. In analog summing, groups or Stems The signal is routed through analog signal paths and then picked up again. The benefit should always be assessed at the same playback level.
Microphones, room acoustics, converters, samples, and the engineer's decisions all influence the result. A serious evaluation avoids blanket percentages and compares specific before-and-after versions.
Why does my song sound quieter on Spotify, YouTube, or Apple Music?
Background to the practical example
The starting point was a client master that, despite high measured loudness, sounded less impactful than reference tracks after being published on YouTube. Analysis showed that not only LUFS (low volume full-range frequencies), but primarily energy distribution, frequency balance, transients, and arrangement determine the perceived sound.
The interview with Andreas Balaskas from the Masterlab Academy was recorded as a NightTalk on November 5, 2018. It remains embedded as a historical practical example; the technical platform specifications in this guide have been updated independently to reflect the status as of September 2026.
Two songs with the same Integrated LUFS value can sound very different in terms of loudness. Crucial factors include arrangement, frequency balance, transients, distortion, saturation, dynamics, and the perceived loudness within the respective frequency range. The platform's measurement setup and disabled normalization also affect the comparison.
Therefore, always evaluate a Master's degree against relevant references. matched playback levelOnly then can you hear whether more limiting actually improves the sound or just makes it louder in the short term.
How loud should I master my song?
Peak-Studios explored this topic in collaboration with Andreas Brede as part of his thesis at Triagon Academy. For this purpose, released productions from various genres were analyzed with regard to their loudness on streaming platforms. The results support a genre- and material-dependent decision rather than a rigid LUFS target.
Genre-related analyses and practical articles
We'll get the best out of your sound.
Streaming mastering: Does my song need to be mastered at -14 LUFS?
Master your song at a volume that makes sense given the sound, genre, and desired dynamics – not automatically at -14 LUFS. Then check how the master sounds after a simulated normalization.
Practice workflow
- Create the final master at the desired loudness.
- Measure Integrated LUFS and True Peak.
- Set the master and reference to the same playback level.
- Compare transients, bass balance, distortion, depth, and dynamics.
- Adjust the limiting or tonality as needed – not just the output control.
For Spotify, you can also simulate the modes -14, -11, and -19 LUFS. These values reflect Spotify settings and are not universal target values for Apple Music, YouTube, or TIDAL.
Do I need to bring my Master's level down to -14 LUFS?
No. -14 LUFS is the reference for Spotify's "Normal" setting, not a universal upload or mastering standard. What matters is that the master sounds good and delivers a convincing performance after level-matched playback.
How is Spotify currently normalizing?
Spotify specifies -14 LUFS for "Normal," -11 LUFS for "Loud," and -19 LUFS for "Quiet." "Normal" and "Quiet" operate without a limiter; "Loud" uses a limiter at -1 dB sample peak. Normalization may be disabled and is not available in every client.
Dolby Atmos on Apple Music
Apple Music offers spatial audio playback with Dolby Atmos at no extra cost as part of its subscription. Atmos masters are not delivered as standard stereo WAV files, but as BWF/ADM files via an authorized distribution partner. Apple's current Asset Guide requires 24-bit audio at 48 or 96 kHz for Atmos; the integrated loudness must not exceed -18 LKFS.
The Atmos master is a separate mix. Simply upscaling a stereo master does not replace a controlled immersive mix or testing of binaural playback.
Lossless Audio on Apple Music and Spotify
Apple Music delivers its catalog as lossless audio in ALAC format; Hi-Resolution Lossless goes up to 24-bit/192 kHz. Apple accepts WAV, FLAC, and ALAC files in 16-bit or 24-bit, with sample rates from 44,1 to 192 kHz. Apple recommends delivering the highest native resolution of the project and avoiding unnecessary upsampling.
Spotify has offered lossless playback for Premium subscriptions since 2025. According to Spotify, loudness normalization also applies to lossless files. Therefore, the file you upload to the distributor should still be a clean, lossless master.
More on this: Spotify Lossless – current overview.
Apple Music Sound Check: no fixed public -16 LUFS mastering target
Sound Check analyzes the loudness and adjusts the playback level between tracks. However, Apple does not publish a universally binding LUFS target value that every music master must meet. The often-cited figure of -16 LUFS is a common practical value derived from measurements, but not a current universal Apple delivery specification.
Sound Check can vary depending on the app, device, and user settings, or may be disabled altogether. Therefore, the master should also work without normalization.
YouTube Music: no reliable, general -7 LUFS requirement
YouTube and YouTube Music adjust loud content during playback, but do not publish a universally binding LUFS upload value for music masters. The "Content loudness" indicator in "Stats for Nerds" describes the level difference that YouTube applies during playback; it is not an upload requirement.
The previously mentioned blanket normalization to -7 LUFS has therefore been removed. For a meaningful assessment of the sound quality, current measurements on the specific upload and a level-matched comparison are recommended.
TIDAL: FLAC for stereo, Dolby Atmos for immersive content
TIDAL removed MQA and Sony 360 Reality Audio from its apps on July 24, 2024. Stereo content is offered in FLAC format; HiRes FLAC supports resolutions up to 24-bit/192kHz, depending on the available source. For immersive playback, TIDAL uses Dolby Atmos.
Technical catalog delivery is handled through the distributor. TIDAL does not currently specify a fixed, generally published LUFS mastering target. Therefore, check your distributor's specifications and evaluate the master both normalized and at its original loudness.
Our system organizes the complete workflow from mixing to streaming delivery. Mastering Guide .
YouTube: Current audio delivery for partners and regular channels
For Content ID and partner deliveries, YouTube recommends lossless audio with 48 kHz / 24 bit16-bit audio is accepted. FLAC and uncompressed PCM are among the supported formats. Eclipsa Audio can handle up to 18 channels in partner workflows.
This specification does not apply to direct audio uploads for every channel. Regular YouTube channels cannot publish a standalone audio file as a video; the audio must be embedded in a supported video container.
Details and practical examples: YouTube Audio Guidelines. With our Loudness Meter Can you check LUFS and True Peak before uploading?
Further reliable information and tools
Is your mix ready for mastering?
Our mix analysis tells you whether your song is ready for mastering and which aspects you can improve yourself in the mix beforehand.
