SplitFire AI Audio Separator vs Low Poly Synth

SplitFire splits any song into stems, right on your iPhone or iPad. Vocals, drums, bass, and more, extracted by AI that runs entirely on your device. No upload, no cloud, no queue. Your audio never leaves your machine. HOW IT WORKS - Pick any audio file: MP3, WAV, FLAC, M4A, AAC, or OGG - Choose an AI model: fast 4-stem, fine-tuned, or 6-stem separation - Get clean stems as WAV files, accelerated by your device's GPU WHAT YOU CAN MAKE - Karaoke tracks: remove the vocals from any song - Practice tracks: mute your instrument and play along - Backing tracks for bass, drums, or guitar - Acapellas, remixes, samples, and DJ edits PRIVATE BY DESIGN - 100 percent on-device processing - Works offline after a one-time model download - No account needed to separate audio SYNCED BY YOUR ICLOUD - Stems save to a SplitFire folder in iCloud Drive and sync across your iPhone, iPad, and Mac - Real files: open them in the Files app or drag them into your DAW on the Mac, no app needed - Sync runs through your own iCloud account, never through our servers - iCloud off? Everything still works, stored locally on your device AND MORE - History: browse everything you have separated on any device, with per-stem sync state - Optional free SplitFire AI account - Made by Splitfire AB in Sweden Support: https://splitfire.ai/contact

Low Poly Synth is a generator of melodic atmospheric sounds with billions of unique timbres. The app is built on modified synthesis algorithms from Fractal Bits, Relic Waves, and Relic Flow, with the addition of pure tone generation code. The sound space is divided into 4294967296 presets, each corresponding to a set of 8 sample textures. The volume of an individual sample can be adjusted in the top panel or via MIDI (Settings -> MIDI Mapping). The bottom panel is a classic musical keyboard that sets the note for the sounds in the set. Each note triggers all 8 samples at once - together they form a complex harmonic pattern that almost never repeats over time due to the different lengths of the looped fragments. Features: * search for new sounds with three buttons: next random set, edit code, previous set; each set of 8 sounds corresponds to a code of 8 characters (you can save/copy/paste it as plain text); * three types of keyboards for live performing: on-screen, PC keyboard, MIDI input; for the on-screen keyboard, you can set the number of displayed octaves and scale; * 20+ parameters can be controlled via MIDI or directly through the app interface: Settings -> MIDI Mapping; * real-time audio recording to WAV (32-bit); * export to: WAV (one file or a set), SunVox (samples + effects in one file), text clipboard; * the LCK button freezes individual samples - they will not change during the search for new sets; * Audio Unit Extension (AUv3), Wi-Fi export/import. Double clicking on a parameter opens a window for setting the exact value. If you enable the Hold option, the notes will play endlessly, without reacting to key release (noteOff) events; turning the note on again works like turning it off; there are two ways to enable this option: 1) using the Hold parameter in the "MIDI Mapping" window; 2) by pressing the HOLD button, which appears in place of LCK while playing the musical keyboard: press HOLD, release the desired notes - then the released notes will continue to play.