The open-source cross-reality audio workstation for spatial sound design, collaborative mixing, and generative music composition.
Pulsewave Studio reimagines the music production landscape as a living, breathing ecosystem—where sound behaves like matter in a physical universe, tracks flow as dynamic waveforms in a zero-gravity workspace, and collaboration happens across time zones asynchronously through a "sound-first" social layer. Born from the vision of an open-band future, this platform decouples music creation from hardware dependency, operating entirely in the browser with a responsive grid that adapts from smartphone to 4K studio displays.
Unlike traditional DAWs that mimic tape machines and mixing consoles, Pulsewave Studio treats every audio project as a sonic organism—with stems that breathe independently, effects that mutate over playback cycles, and a spatial audio engine that places instruments in a 360-degree environment. The platform is built for bedroom producers, live-looping performers, sound therapists, and field recordists who need a tool that evolves with their artistic process rather than constraining it.
Pulsewave Studio eliminates the friction between inspiration and capture. The entire interface loads under two seconds on a standard 5G connection, with offline-ready service workers caching core assets for mobile production on flights or in remote locations. The session format is an open JSON schema (.pulse files) that can be version-controlled with Git, shared as a URL, or exported as a self-contained HTML document that plays back in any modern browser without dependencies.
The architecture follows a modular node-graph philosophy: every track is a chain of processors that can be reordered, bypassed, or swapped in real-time without interrupting audio flow. The mixer section uses a novel "gravity" metaphor—louder channels appear larger and drift toward the center, quieter ones orbit at the periphery. This visual feedback makes balancing a 50-track orchestral session as intuitive as observing a planetary system.
- Unlimited audio and MIDI tracks with per-clip automation envelopes
- Non-destructive comping with "take lanes" that stack vertically, inspired by code branching
- Time-stretching and pitch-shifting powered by a custom phase vocoder engine
- Groove quantize with humanization curves that feel organic, not robotic
- Convolution-based cabinet emulator with 200+ impulse responses (IRs) from vintage amps
- Drag-and-drop pedal chain with over 60 effects: fuzz, delay, reverb, modulation, pitch shift
- Real-time parameter modulation via LFOs, envelope followers, or MIDI controllers
- "Amp Room" 3D visualization that maps tone stack settings to physical speaker cone movement
- On-device machine learning separating vocals, drums, bass, and other instruments
- Spectral view with "paintbrush" selection: isolate or remove specific frequencies like erasing noise
- Source separation runs entirely in WebAssembly—no data leaves your computer
- Public "Sound Garden" where users share loop packs, preset chains, and full sessions
- Asynchronous collaboration: record a guitar part, tag a collaborator, they receive a MIDI slice to respond
- Comment on specific bars and beats—like code review, but for music arrangements
- Adaptive layout: single-column on mobile, multi-viewport on desktop with floating panels
- Touch gestures for iOS/Android: pinch-zoom waveforms, swipe to switch track views
- PWA installable with offline storage—production anywhere, even on an airplane
- Interface localized in 18 languages, including right-to-left script support
- High-contrast mode and screen-reader-optimized track widgets
- Colorblind-friendly waveform palettes with pattern overlays instead of color-only indicators
- Community-driven support via integrated forum threads inside the DAW sidebar
- AI-assisted bug reporting that automatically captures session state and system specs
- Monthly "Office Hours" live stream where core contributors answer questions and demo features
- Open Pulsewave Studio in any modern browser (Chrome, Firefox, Safari, Edge—v2026+ recommended).
- Click "New Session" and choose a template: "4-Track Beat," "Orchestral Sketch," "Field Recording Canvas," or "Blank Slate."
- Record or import audio by dragging WAV/MP3/FLAC files directly onto the waveform grid.
- Right-click a track to add effects from the pedalboard library—search by category or sound character (e.g., "warm," "glassy," "textured").
- Press the spacebar to audition your mix. Adjust track volumes by dragging the "gravity" orbs near each fader.
- Share your session as a public link with a single click—collaborators can open it in their own Pulsewave instance without needing an account.
No account? No problem. Sessions are stored locally by default. Cloud sync is optional and encrypted end-to-end using Web Crypto API.
Pulsewave Studio is built on a micro-frontend architecture. Each core feature lives in its own repository under the @pulsewave namespace:
| Package | Description |
|---|---|
@pulsewave/engine |
Real-time audio graph, WASM DSP, scheduling |
@pulsewave/ui-canvas |
Waveform rendering, spectrogram, touch handling |
@pulsewave/pedalboard |
Effect chain modeling, IR loader, Web Audio nodes |
@pulsewave/social-core |
ActivityPub-compatible feed, session versioning |
@pulsewave/spectral |
On-device ML inference for source separation |
The engine uses a shared-array-buffer-free architecture to ensure compatibility with all major browsers, including Safari's strict memory model. Audio rendering is double-buffered with requestAnimationFrame sync for sub-10ms latency.
We welcome contributions of all kinds—from bug fixes and new effects to translation improvements and accessibility audits. Here's how you can get involved:
- Submit a pull request with a clear description of the change and a recorded audio example if adding a new effect.
- Report issues with a minimal
.pulsesession file attached to reproduce the bug. - Contribute translations via our localization platform (link in repository wiki).
- Write documentation—every public API function should have at least one usage example.
Code style follows the Pulse Standard: 2-space indentation, meaningful variable names (no single-letter variables except loop indices), and comments that explain why, not what.
This project is licensed under the MIT License—see the LICENSE file for full terms. You are free to use, modify, and distribute Pulsewave Studio for any purpose, including commercial projects, as long as the original copyright notice is retained.
Trademark Notice: "Pulsewave Studio" and the waveform-logo mark are trademarks of the Pulsewave collective. Use of these marks for derivative projects must include attribution.
Pulsewave Studio is provided "as is," without warranty of any kind, express or implied. The developers are not responsible for any loss of audio data, corrupted sessions, or emotional distress caused by the creative process. Audio files generated by this software may be subject to copyright laws in your jurisdiction—please ensure you own the rights to any samples or recordings you use.
The machine learning stem separation feature is intended for remixing, analysis, and educational purposes. Users are responsible for complying with applicable copyright laws when using separated stems.
This project is not affiliated with any commercial DAW manufacturer, guitar amplifier company, or cloud service provider. All brand names and product names mentioned in documentation are for reference purposes only.