Freesound Forums

Hi everyone,

This post is to let you know about something I've been working on lately which might be of your interest It is a hardware music sampler which connects to Freesound and allows to search and play (and attribute!) sounds. I call it "SOURCE". Here is some info about SOURCE copied from the source code repository page:

---

SOURCE is an open-source music sampler powered by Freesound's collection of 500k Creative Commons sounds contributed by a community of thousands of people around the world. SOURCE is a sampler that does not sample. Instead, it provides different ways to load sounds from Freesound and instantly generate new sound palettes to enrich the creative process and bring an endless SOURCE of inspiration.

SOURCE is designed to run as a stand-alone device on a hardware solution based on a Raspberry Pi 4, the Elk Pi hat for the Raspberry Pi (which provides low-latency multi-channel audio I/O), and the Elk BLACKBOARD controller board (which provides the user interface elements including buttons, faders, a display, and the audio I/O connectors). However, the core of SOURCE is implemented as a standard audio plugin using JUCE. That allows SOURCE to also be loaded in DAWs that support VST/AU plugins, or even run as a stand-alone and cross-platform application in desktop computers (eventhough with somewhat limited functionality).

Here is a video demonstrating some features of SOURCE in action: https://www.youtube.com/watch?v=7EXMY0AvBxo

More information in the source code repository

I hope you like it!!

Wow, very impressive. Thank you for sharing.
Along the video you also showed how differently the search on freesound can be used-I haven't used it that much yet and was surprised about the possibilities.
So, did I understand you correctly, that SOURCE could be used with any MIDI, like e.g. Komplete Kontrol Keyboards?

Hi, thanks for your words. Yes, SOURCE responds to MIDI messages like any other synthetizer/sampler.

This is so cool. Very impressed!

Just thinking out loud here..

Could the CV input be used to select the sound directly from the server? With a fast enough connection and the right sounds latency would be tolerable I'd have thought.

What if it could sample? So the user enters attribution info before a session and then it operates like a traditional sampler with the option to upload to Freesound at the push of a button. Maybe with some sort AI "middleman" handling the tagging as to not disrupt the users flow. It would open up some interesting collaboration possibilities too..

Hi modularsamples,,

CV control would be great and should be technically possible id modulating at slow speeds, or when sounds have already been downloaded.

Sampling and uploading to Freesound would also be a natural next step, you're right! Hopefully at some point I'lll be able to invest more efforts in SOURCE and implement some of that stuff.