ABSeq is for comparing finished audio files. Producers, mix and mastering engineers, artists, labels, beatmakers and students use it to pick between versions of a song: two mixes, a mix and its master, several masters, or their track against a commercial reference. It plays up to four files in sync on iPhone or iPad, matches their loudness with Auto Gain, shows LUFS and true peak for each, and can hide which file is which in a blind test. For a feature-by-feature walk-through, read how ABSeq works.
Choosing between two mixes
You bounced mix 6 last night and mix 7 this morning. Mix 7 has the vocal up 1 dB and a brighter snare. You can't tell any more if it is better or just different.
Load mix 6 into A and mix 7 into B, switch on Auto Gain and press Play. Both loop in sync, so pressing A/B on the second chorus takes you to the second chorus of the other mix. Auto Gain turns the louder file down to the quieter one, so a 1 dB level change in the bounce doesn't make one mix win by being louder. If you still can't decide, run a blind test of 10 rounds and see which one you pick.
Checking a master against the mix
A master usually comes back several dB louder than the mix. Played as they are, the master sounds bigger almost every time, and that tells you nothing about what the mastering did.
Load the approved mix into A and the master into B. Auto Gain brings the master down to the mix's integrated loudness. Now you can hear whether the low end got tighter, whether the vocal still sits where you put it, and whether the limiter flattened the drums. The tiles show both files' LUFS and true peak, so you also see the master's real loudness and whether it peaks above −1 dBTP. The A/B difference graph (Pro) shows band by band where the master has more or less than the mix. The guide to comparing masters goes further.
Picking between several masters
Sometimes you have more than one master of the same mix: from two mastering engineers, from an engineer and an online or AI mastering service, or from one service at different settings. Each one is likely to be at a different loudness, so the loudest tends to win a casual listen.
Load up to four masters into A, B, C and D. With three or more tracks the A/B key reads STEP and moves through them in turn, and Auto Gain brings all of them down to the quietest. Turn on the sequencer at the song's tempo with a division of 1/1, and ABSeq switches every bar while you listen. At 120 BPM that is every 2 seconds. Narrow the four to two, then run a blind test on the last pair.
Referencing a commercial track
You are mixing a song and want to know if your low end is in the same area as a release you like in the same style. Buy or export a good-quality file of the reference and load it into B next to your mix in A.
At matched loudness, switching back and forth shows where your mix is thinner, darker or busier than the reference. The free spectrum shows the track you hear as bars and the other as a line across 39 bands from 25 Hz to 20 kHz. Bands below about 125 Hz are coarse, so trust your ears more than the graph down there. More on picking and using references in the guide to reference tracks.
The car and earbud test
A mix that sounds right on studio monitors can lose its bass in earbuds or turn boomy in the car. You want to hear both versions on the system where most people will play them.
ABSeq keeps playing and switching with the screen locked. Connect to the car over Bluetooth or a cable, load the two mixes, press Play and lock the phone. The next-track control on the Lock Screen or in Control Center moves to the next version, and the sequencer can switch for you so you keep your hands on the wheel. The same works with AirPods, wired earbuds or a Bluetooth speaker. See the car test guide for what to listen for.
Checking streaming loudness before release
Before you upload, you want to know how the services will treat your master. Spotify normalizes to −14 LUFS by default. A master at −8.5 LUFS gets turned down about 5.5 dB, and it then plays at the same level as a quieter master, with less dynamic range left.
Load the master and read its tile. The LUFS figure is the integrated loudness of the whole file, and the dBTP figure turns red above −1 dBTP. Tap the readout to see how much Spotify, Apple Music, YouTube, Tidal, Amazon Music and Deezer would change the level. "Spotify −1.8 dB" means Spotify would turn it down 1.8 dB. Only Spotify publishes its rules in full, so the other figures are estimates. The guide to LUFS for streaming lists each service.
Artists and labels approving versions
An artist gets three revisions of a mix by email and has to reply with one. A label A&R has the radio edit and the album version to sign off. Neither wants to open a DAW.
Save the files from Mail to the Files app, load them into ABSeq and switch between them at matched loudness. A blind test with the file names hidden until the end gives a decision that isn't swayed by "v3_FINAL". Save the comparison as a session, so you can reopen the same files and results later. The free version keeps 3 saved sessions. With Pro, sessions and their audio sync over your own iCloud to your other iPhones and iPads.
Beatmakers comparing bounces
You made the same beat with two different kick samples, or bounced it once with a clipper on the master and once without. You want to hear the difference in the loop, at the same point.
Load both bounces. They loop together, so switching keeps the groove going. ABSeq detects the tempo in each file, and the sequencer can use it to switch on every beat division from 1/32 to a whole bar. A fast division such as 1/4 makes small differences in a kick's attack easier to hear. Because Auto Gain never turns a track up, nothing clips when you compare.
Ear training with blind tests
Students and anyone training their ears can test what they really hear. Export the same mix twice, with one change: a 1 dB shelf boost at 10 kHz, a different reverb, a WAV against an MP3. Then run a blind test.
Blind Test plays the same short passage from both files each round, shuffled and labelled Version A and Version B. You pick the one you prefer, for 1 to 50 rounds, with the seconds per passage you set. If you pick the same file 9 times in 10, that is unlikely to be chance. A 5 and 5 split means you can't reliably hear the change. It is a preference test, not an ABX identification test. The guide to blind A/B tests explains how many rounds to run.
What ABSeq is not
ABSeq is a player and a meter. It doesn't edit, mix or master audio, and it doesn't write anything back to your files. Auto Gain only lowers the playback level inside the app.
It is not a DAW, so there are no tracks to record, no plugins and no export. It is not a plugin either, so it doesn't run inside your DAW. You bounce your versions from the DAW, put them in the Files app, and compare them in ABSeq. It runs on iPhone and iPad with iOS or iPadOS 26 or later, and opens WAV, AIFF, AIFC, CAF, FLAC, MP3, M4A and AAC.
Questions
What is an A/B comparison app used for?
An A/B comparison app plays two or more versions of a song in sync so you can switch between them at the same moment and pick the better one. Producers use it for mixes, masters and reference tracks, ideally at matched loudness so the louder version doesn't win by volume alone.
Can I compare AI mastering with a mastering engineer's master?
Yes. Load both masters, and up to two more, into ABSeq. Auto Gain turns them all down to the quietest, so you compare the sound and not the loudness, and a blind test hides which master is which until the results.
Is ABSeq a DAW or a plugin?
Neither. ABSeq is a standalone iPhone and iPad app that plays and measures finished audio files. It doesn't edit or master audio and doesn't run inside a DAW.
How do I check if my master is loud enough for Spotify?
Measure its integrated LUFS and true peak. Spotify plays tracks at −14 LUFS by default and turns louder ones down, so a −9 LUFS master is turned down 5 dB. In ABSeq, the tile shows LUFS and dBTP, and tapping it shows how much each service would change the level.