This version is designed for a computer. Vocal Stem additionally downloads a model of about 289 MB and needs roughly 2 GB of free memory. For a short trim, use the audio cutter.
ℹ️ Only Vocal Stem needs the large model: about 289 MB on first use and roughly 2 GB of free memory. BPM Shift, Granular and Pitch Shift use other processing methods. Runtime depends on the method, file length and computer.
⚡ Trim a short excerpt first to judge the result without a long wait. For Suno uploads, check your plan’s limit. A shorter file alone does not resolve a recording-match rejection.
Make a modified version of your recording and compare it with the original. Choose pitch shifting, slowing down, granular processing or vocal-stem processing. Audio is processed in your browser and is not sent to our server.
For a failed upload, start with the Suno error checklist. Changing the sound does not fix every error or guarantee acceptance.
1. Upload audio (mp3 / wav / m4a / flac, up to 50 MB)
🎵
Drop a file or click to choose
mp3 · wav · m4a · flac · ogg
📂
—
—
2. Processing method
🎚️ Vocal Stem separates and processes the vocal. It is the most demanding mode; a long recording may take several minutes. Listen for changes to consonants and phrase boundaries.
🚀 BPM Shift slows the recording and attenuates part of the vocal range. Strong settings noticeably change rhythm and intelligibility. A slower reference may also affect the tempo of a new generation.
🔬 Granular rearranges short fragments and varies their pitch. Overall duration is retained, but attacks and vocals can become shaky. Start with a light setting and check whether the phrase still sounds usable.
🎼 Pitch Shift changes pitch without intentionally changing tempo. Use it to compare a reference in another register. Vocals may develop artifacts; compare the formant-preserving option with the standard result.
3. Anonymization strength
💡 Start with Moderate. If the filter rejects — try Subtle (counterintuitive but sometimes works better) or Aggressive. The most recognisable global hits may not pass on any preset — try BPM Shift method (it's stronger against lyrics).
💡 Start with Medium. Stronger preset = muddier vocal — that's the point: if Suno can't parse the lyrics, it can't identify the song. Strong nearly kills intelligibility — good for the stickiest hits. Output sounds «like from a barrel» — that's expected, Suno will only pick up the melody.
💡 Start with Medium. Light keeps the sound near-original, Strong makes the track noticeably «jittery» — shorter grains and wider shuffle window. This is the nuclear option against fingerprint check. Output sounds «jittery» but the melody survives — Suno will pick it up and generate a clean track.
💡 A single preset — +2 semitones −20 cents (the exact Adobe Audition settings). Tempo untouched, minimal artifacts. If it doesn't pass the filter — try BPM Shift or Granular (they're more aggressive).
💡 Listen to the beginning, chorus and ending. If there are dropouts or excessive artifacts, use a milder setting. Processing is complete; that does not mean Suno will accept the upload or restore the original quality.
🎚 A slowed reference can lead to a slower generation. Measure its tempo before correcting it: one fixed speed-up percentage will not fit every result. Heavy time-stretching affects sound quality.
🎵 Step 2 — shape the actual sound
Your file is ready. Save it and listen back. For your next song, describe the instruments, vocal character and chorus dynamics in the free prompt builder.
Start with free genre examples or explore a direction in the artist catalog. See pricing for full access details.
🛡️ Legal note:
This tool is for audio you have rights to — your own recordings, public-domain material, licensed content. It doesn't grant rights you don't already have.
Audio’s clean — now make Suno nail the style
Ready style prompts for 1166 artists — genre, instrument and vocal descriptions for your next track.