Support
Stem Splitter — Vocal Remover & Karaoke Maker
Getting Started
How do I split a song?
Open the app and tap the + button in the Library. Pick an audio file from Files, iCloud Drive, AirDrop, or share a track from another app via the iOS share sheet. Separation kicks off automatically; the result screen shows four stem tracks with mute, solo, and volume per stem.
What stems do I get?
Four: Vocals, Drums, Bass, and Other (which contains everything else — guitars, keys, synths, harmonic instruments). All four are produced for every separation, free or Premium.
How do I make a karaoke / instrumental track?
After separation, open the project, tap the speaker icon on the Vocals row to mute it, and play. To export the instrumental as a single file, tap Export and choose Custom Mix (Premium feature).
What audio formats are supported?
MP3, M4A, AAC, WAV, AIFF, and FLAC. Files are decoded and resampled to the model’s working rate during separation. Export formats are MP3 192 kbps (free), MP3 320 kbps, and WAV 24-bit (Premium).
Quality & Performance
How long does separation take?
Roughly 1× to 2× real time on iPhone 13 Pro and newer. A 4-minute song separates in 4–8 minutes on the iPhone 13 Pro, faster on iPhone 15 Pro / 16 Pro / 17 series. Older devices (iPhone 12 / 12 Pro) may take 2–3× real time.
I hear bleeding between stems. Is that a bug?
Some bleed is expected — especially on the Other stem — and is a known limitation of all source-separation models, on-device or cloud. Bleed is most noticeable when the source is heavily compressed, low-bitrate, or has complex overlapping instrumentation. For best results, start from the highest-quality source you have (lossless > 320 MP3 > 192 MP3).
Does it work on classical, jazz, or voice-only recordings?
The model is trained primarily on popular-music mixes (rock, pop, hip-hop, electronic). Classical music, jazz with sparse instrumentation, and voice-only recordings (no music to separate from) often produce poor results. For voice-only audio, you may want a different tool — try Bulpara’s Dereverb for echo removal or Aside for transcription.
What devices are supported?
iPhone 12 (A14) and later, and iPad with M-series chips for the fastest experience. Older devices lack the Neural Engine throughput needed for on-device separation.
Do I need an internet connection?
Only for the one-time model download on first launch (around 100–300 MB). After that, separation runs entirely on your device, even on a plane. Toggle Airplane Mode after the first separation to confirm.
Apple Music & DRM
Why can’t I import a song from Apple Music?
Apple Music tracks are DRM-protected and cannot be exported by any third-party app. The App rejects them at import with a clear message. Bring your own files — tracks from your music library that you own (purchased from iTunes long ago, ripped from CDs, or files you produced) work fine. Files imported via the share sheet from Voice Memos, Files, or AirDrop are unaffected.
What about Spotify, YouTube, or SoundCloud?
The App can’t download audio from streaming services — that’s a separate problem (and often a Terms-of-Service issue with each service). If you have audio files saved locally that you have the right to process, drop them in.
Free vs Premium
What’s free vs Premium?
Free: 3 separations per month, all four stems included, MP3 192 kbps export, up to 6 minutes per song, with banner and post-export interstitial ads. Premium: Unlimited separations, no length cap, WAV and MP3 320 kbps export, custom mute/solo mix downloads, background processing for long tracks, and zero ads.
Is there a free trial?
Yes — the Premium Annual plan ($29.99/year) includes a 7-day free trial. You’ll only be charged at the end of the trial. Cancel anytime in your Apple ID subscription settings before the trial ends.
How do I restore my purchase?
Open Settings inside the App and tap "Restore Purchases". This re-applies your active subscription if you’ve previously purchased one on this Apple ID.
How do I cancel my subscription?
Open the iOS Settings app, tap your name at the top, then tap Subscriptions. Find Stem Splitter and tap Cancel Subscription. Cancellation takes effect at the end of the current billing period.
Privacy
Where do my audio files go?
They stay on your iPhone. The App does not upload audio to any server. Separation runs on-device using a Core ML model that runs on the Apple Neural Engine. You can confirm by toggling Airplane Mode — separation still works after the one-time model download. See the Privacy Policy for full details.
What permissions does the app need?
Files / iCloud Drive (only for the audio file you pick) and Notifications (optional, for background-processing completion alerts). The App does not request microphone, photos, contacts, or location access.
Contact Us
Email Support
For questions, bug reports, or feature requests:
We typically respond within 48 hours. If you’re reporting a bug, sending a sample track and the device model helps us reproduce.