INTERACTIVE MIXER / DEVELOPERS & DEEP WORK
Ambient Sound Mixer for Coding & Deep Work
Voice-free by default so human speech never competes with code reading or debugging. Balance steady rain, soft mechanical keyboard taps, wind, stream, and fire straight in your browser — with café voices kept at 0% until you explicitly choose to bring them in.
Every track here is free to play with no sign-up. Drag any slider to tune the balance live, or replace a track with your own local file on Sources & Licences.
Voice-free by default
Café Voices starts at 0% to avoid verbal working-memory interference while coding
Zero install, lightweight tab
Runs quietly in a background browser tab with no Electron app eating RAM
Shareable team/desk presets
Encode your exact focus balance into a /mix/SYL-… URL for one-click reopen
Why lyrics and speech hurt debugging more than rain
Reading stack traces, naming abstractions, and holding state in working memory rely heavily on your phonological loop. Even quiet songs with vocals or clear coffee-shop chatter compete for that same mental channel. Steady broadband textures — rain, stream, distant wind, and subtle keyboard rhythm — mask open-office noise without introducing words your brain has to parse.
Switch modes without switching apps
Keep Café Voices at 0% during heavy debugging or architecture work. When you move to writing documentation, pull-request descriptions, or Slack updates, nudge Café Voices up to 25–35% or click the "Light Café Drafting" preset above — all inside the same browser tab.
Pair with a calm second-monitor scene or MCP agent
Want a quiet visual anchor on a second monitor? Open the Chiang Mai Focus Soundscape (free, with café speech off by default). If you use Claude, Cursor, or another MCP-compatible coding assistant, you can also connect Sound Cocktail’s read-only MCP endpoint (/api/mcp) to generate playable focus mix links directly from your IDE chat.
Frequently asked questions
Is this coding sound mixer completely free of vocals and speech?
By default, yes: Rain, Keyboard, Wind, Stream, and Fire contain zero human speech, and Café Voices starts at 0%. You only hear café chatter if you drag that slider up.
Can my AI coding assistant (Cursor / Claude) generate Sound Cocktail mix links?
Yes. Sound Cocktail exposes a read-only MCP endpoint at https://soundcocktail.app/api/mcp with list_sounds, recommend_soundscape, and build_mix_url tools.