About Sonicaura Studio
Sonicaura Studio builds AI audio tools that run in your browser and on our servers — and that can be called directly from AI assistants through the Model Context Protocol (MCP). Three core tools: stem separation, AI denoise, and spatial reverb.
Why we built it
Audio tools have historically been locked inside expensive desktop software. If you wanted to split a song into stems, remove hum from a podcast recording, or add reverb to a dry vocal, you either paid for a DAW plugin or learned a complicated interface first.
We thought that was unnecessary. Modern AI models can do most of this work in seconds — the hard part is making them accessible: no install, no account, no friction. So we built Sonicaura Studio around three principles:
- →Runs where you are. Simple tasks run in your browser; heavy AI models run on our servers. You do not need to know which is which.
- →Measurable results. Every AI tool includes an A/B comparison so you can hear the difference, and most include a live frequency spectrum so you can see it too.
- →Works with the tools you already use. All three AI tools are callable through the Model Context Protocol (MCP), so AI assistants such as Claude and Cursor can run them directly on your behalf.
What we make
AI Stem Separator
Split any song into 2, 4, or 6 stems — vocals, drums, bass, guitar, piano — with automatic chord recognition.
AI Sound Denoise
Remove hiss, hum, static and room tone while preserving instruments and vocals. Includes a live frequency spectrum.
AI Sound Reverb
Add vocal booth, studio, bathroom, hall or cathedral reverb to any recording — five presets, one click.
Call Sonicaura from Claude and Cursor via MCP
All three AI tools are exposed through the Model Context Protocol (MCP)— an open standard that lets AI assistants use external tools on your behalf. Instead of clicking through a web interface, you describe what you want in chat, and Claude Desktop, Cursor, or any other MCP-compatible client runs it for you and returns a signed download link.
Natural language. Ask your assistant to "split this song into stems" or "remove the background noise from this recording".
One-time setup. Add the MCP server URL and your API key. Setup takes a few minutes and is documented for each client.
Same tools, same credits. MCP calls use the same credit balance as the web interface — one credit per task, no separate pricing.
How it works
Free Tools for Every Practice Session
Warm up, tune up and train your ear — 15 browser-based audio tools, no download or install required.
Audio Editor & Trimmer
Trim, cut, fade, reverse and export audio right in your browser. Drag on the waveform to select, edit with one click, and download as WAV — nothing is uploaded.
Browser Vocal Denoise
Free browser-based vocal denoiser with noise gate, compressor and EQ. Runs entirely on your device — nothing is uploaded.
Browser Sound Reverb
Add room, hall, or cathedral reverb to any recording — with real-time waveform, A/B comparison, and one-click WAV export. Runs entirely in your browser.
Chord Practice
Import a Chord-Pro chord sheet and practise along a synced timeline with live fingering diagrams.
Chord Finder
Look up any of 34 guitar chords and compare open, barre and high-register fingerings on an interactive fretboard.
Format Converter
Turn .m4a and .mp4 files into clean MP3 or WAV before processing.
Pitch & Tuner Test
Sing into your mic and get a real-time pitch accuracy score.
Vocal Range Finder
Find your lowest and highest notes, and your exact voice type.
Ear Training
Train your ear to tell apart subtle microtonal pitch differences.
Frequently asked questions
What does Sonicaura Studio do?
Sonicaura Studio builds AI audio tools that run in your browser and on our servers: stem separation, AI denoise, and spatial reverb. All three are designed for musicians, podcasters, and creators who want professional results without installing a DAW plugin.
Can I use Sonicaura Studio from Claude or Cursor?
Yes. All three AI tools are exposed through the Model Context Protocol (MCP), which means AI assistants such as Claude Desktop and Cursor can call them directly. You can ask the assistant in plain language — "split this song into stems", "remove the background noise from this recording", "add cathedral reverb" — and it will run the task on your behalf and return a signed download link.
What is MCP and why does it matter for audio tools?
MCP (Model Context Protocol) is an open standard that lets AI assistants use external tools on your behalf. Instead of clicking through a web interface, you describe what you want in chat and the assistant calls the right tool. For audio, this means you can process a file without leaving Claude or Cursor — useful for podcast edits, quick demos, or batch work.
How do I connect Sonicaura Studio to my AI assistant?
See the MCP integration page for step-by-step setup. You will need an API key from your Sonicaura account and the MCP server URL. The page includes instructions for Claude Desktop, Cursor, VS Code and other MCP-compatible clients.
Do I need to install any software?
No. Every tool runs in your browser or on our servers. There is nothing to download, no plugin to install, and no account required to try the free browser tools.
How much does it cost?
New accounts get 2 free credits on signup, no credit card required. Each AI task costs 1 credit, regardless of file length, and credits never expire. The browser-based tools (Format Converter, Metronome, Piano, Chord Practice, and others) are free.
Where is my audio stored?
Browser-based tools process audio entirely on your device — nothing is uploaded. Cloud AI tools upload your file to private object storage, process it, and return a signed download link. Files are never shared with third parties.
Who is Sonicaura Studio for?
Musicians practising or remixing songs, podcasters and voiceover artists cleaning up recordings, and creators who need quick audio utilities without opening a DAW. The tools are designed to be useful on the first click, with advanced controls available when you need them.
Get in touch
Questions, feedback, or something not working the way you expect? We read every message.
Contact us