Free tier: up to 3 splits per day for files up to 2 minutes and 50 MB. Paid credits: Starter $4.90 (20 minutes), Plus $9.90 (50 minutes), Pro $19.00 (120 minutes). Credits never expire and are billed per minute used (rounded up).
Key features
Synthesize sounds from oscillators and noise layers
Stack multiple layers with individual pitch and envelope controls
Adjust wave shape: sine, triangle, saw, square or noise
Real-time visualization of sound waves
Apply effects: reverb, chamber, echo, chorus, phone filter and crunch
Share sounds via URL link carrying full sound data
Download sounds as MP3 files
Connect with Claude, ChatGPT, Cursor and VS Code via MCP server
Save sounds to personal browser library
No account or installation required
No signup required for free tier
In-browser stem preview before downloading
Supports MP3, WAV, M4A, and AAC formats
Automatic file deletion after approximately one hour
Usage metered fairly by processing time
Demucs htdemucs AI model for source separation
Privacy-minded with no account library
Free tier up to 2 minutes with 3 daily splits per IP
What makes it different
AI assistant integration through MCP server for natural language sound design
Synthesized from scratch with no samples or recordings
Real-time editing with live audio preview while adjusting
Works as notification sounds, ringtones, and in apps without conversion
No account or signup required to remove vocals
Vocal and instrumental stems remain perfectly aligned for easy switching
Fast processing with no desktop installation or configuration needed
Truly free tier with no hidden checkout or trial card required
Notify Sound
Get Vocal Remover