MIDI SoundFont/DLS player and editor written in TypeScript.
SpessaSynth is a SoundFont2-based real-time synthesizer written in TypeScript, previously pure JavaScript.
This is a repository for the web application, it used to be a monorepo containing the entire project.
SpessaSynth Project index
SpessaSynth Web App
The complete GUI for spessasynth_lib, allowing remixing, experimenting, and playing MIDIs in the coolest way possible.
Online Version (no download needed!)
Download Local Edition
Made with SpessaSynth
Usage Guide
Supported Browsers
Chromium-based: such as Google Chrome, Microsoft Edge, Brave, Opera, etc.
Gecko-based: such as Mozilla Firefox, Floorp, Zen, Waterfox, Librewolf, etc.
Chrome for Android/Chromebook
Firefox for Android
Features
Visualization of the played sequence: with cool effects like visual pitch bend and note-on effects!
Playable keyboard with various sizes: mobile friendly!
Integrated controller for the synthesizer with numerous options: Edit instruments, controllers, and more!
MIDI Lyrics support: Karaoke!
Kar Lyrics support: Karaoke but fancy!
ASS Subtitle support: Karaoke but even fancier!
Music player mode: with support for album covers in .rmi files!
Mobile-friendly UI: Enjoy SpessaSynth on your phone!
Extensive language support:
English
Polish
Japanese (translated by src3453)
French (translated by Davy Triponney & Samuel Louf)
Portuguese (translated by Lucas Gabriel)
Simplified Chinese (By 懒洋君)
Italian (By ItsFilyy)
Russian (By TGaming03)
Georgian (By TGaming03)
Custom locale management system, created specifically for this project
Description of buttons: Hover over the buttons to see what they do!
Web MIDI API support: Use your physical MIDI devices!
WebMidiLink support
Numerous exporting options:
Render the MIDI file (modified or unmodified) to .wav
Render each channel as a separate .wav file
Export the modified MIDI file to .mid
Export the trimmed SoundFont to .sf2
Export a DLS file to .sf2
Export an SF2 file to .dls
Or compress it as .sf3!
Bundle both as .rmi with metadata such as album cover!
Comes bundled with a compressed GeneralUser GS SoundFont to get you started
Still not convinced? Check out these screenshots:
Local Edition
SpessaSynth: Local Edition comes with easier soundfont management than the demo version,
also allowing to use it offline! You can download it here.
[!IMPORTANT]
Firefox is recommended due to Chromium's 4GB memory limit.
A decent computer is also recommended for high-polyphony sound banks.
Recommended high-quality SoundFont for Local Edition
Building from source
bash 复制
git clone "https://github.com/spessasus/SpessaSynth" --depth 1
npm i
npm startThe "start" will automatically build the script and then invoke server.js.
Then it will only invoke server.js so you need to npm run build to rebuild.
The build files:
dist: the online demo along with the bundled Local edition ZIP for downloading. This is what's hosted on GH pages.
local-dev: Local Edition sources with sourcemaps. For development.
server: the Local Edition server. It is here so it can use the root "soundfonts" folder. This is what npm start invokes.
If you would like to help translate SpessaSynth, please read this guide (and thank you!)
If you like this project, consider giving it a star. It really helps out!
Special Thanks
FluidSynth - for the source code that helped implement functionality and fixes
Polyphone - for the soundfont testing and editing tool
Meltysynth - for the initial low-pass filter implementation
RecordingBlogs - for detailed explanations on MIDI messages
stbvorbis.js - for the Vorbis decoder
sl-web-ogg - for the Ogg Vorbis encoder
Bootstrap Icons - for the icons
fflate - for the MIT DEFLATE implementation
JSZip - for the ZIP creation library
foo_midi - for useful resources on XMF file format
Composers - for the demo songs
Falcosoft - for help with the RMIDI format
Christian Collins - for the bundled GeneralUserGS soundfont and various bug reports regarding the synthesizer
And You! - for checking out this project. I hope you like it :)
Major Releases
0.0.1 - Initial release
1.0.0 - Removed the 2-sample limit per voice
2.0.0 - Implemented the worklet system and added support for modulators
3.0.0 - Moved to audioWorkletGlobalScope for offline audio rendering
4.0.0 - Rewrote the project in TypeScript
License
Copyright © 2026 Spessasus
Licensed under the Apache-2.0 License.
Legal
This project is in no way endorsed or otherwise affiliated with the MIDI Manufacturers Association,
Roland Corporation, Yamaha Corporation, Creative Technology Ltd. or E-mu Systems, Inc.,
or any other organization mentioned.
SoundFont® is a registered trademark of Creative Technology Ltd.
All other trademarks are the property of their respective owners.