Orchestron visual CSound platform

Developers
#1

Hi,
I tested the capabilities of openAI Codex and created the following app within a few days.
Feel free to try it out and give feedback :slight_smile:

Repo: GitHub - thomassresearch/orchestron: Orchestron is a visual instrument workshop built on the power of Csound — not a preset browser, not a plugin rack, not a timeline with tracks stacked like office paperwork. You design sound from signal flow upward. Oscillators, envelopes, filters, physical models, control signals — connect them, and the instrument comes alive while you hear it.

Have fun,
Thomas

#2

hi thomas -

this looks amazing, and i wanted to try it out.
i am on debian trixie, and was following your description at

i think you need to add some dependencies under Requirements. i had to
install uv, node and more. perhaps something is still missing; when i run
uv sync and then
make i get:

cd frontend && npm run build

visualcsound-frontend@0.1.0 build
tsc && vite build
src/App.tsx:1:67 - error TS2307: Cannot find module ‘react’ or its
corresponding type declarations.

and many more (Found 1285 errors in 13 files.)
i installed react via
npm create vite@latest my-app – --template react
as described at

best -
joachim

#3

Hi Joachim

Glad you like it. I changed a lot during this weekend, so don’t forget to “git pull”, I fixed lots of bugs, added multilingual support and new features like the „controller sequencer“, but there’s still a lot to do, not all opcodes work and some show an incorrect documentation (when pressing the “?” in the opcode in the graph editor)

I wasn’t aware of missing dependencies, as Codex did a lot of the heavy lifting for me :slight_smile:

What did you have to run besides:
npm create vite@latest my-app – --template react

Did you finally get it to work on Linux? (I’m on MacOS)

Then maybe I should start platform specific infos on how to build and run it. I just added a QUICKSTART.md that give some info, maybe that helps.

Regards,

Thomas

#4

Following up myself: I added support for containerization using Docker. In that operation mode, the realtime audio is streamed back to the browser using WebRTC.
You need Docker, then run:

docker compose build
docker compose up

Then open http://localhost:8000/client with your browser.
There are some latency issues and small periodic glitches still unresolved… to be analyzed :slight_smile:

Thomas

#5

Hi Joachim,

Could you try on your Linux machine: clone the repo, cd into the its folder, then:

sudo apt update
sudo apt install -y nodejs npm alsa-utils
sudo apt install -y --no-install-recommends \
  csound \
  build-essential \
  pkg-config \
  cmake \
  libasound2-dev \
  libjack-jackd2-dev

curl -LsSf https://astral.sh/uv/install.sh | sudo env UV_INSTALL_DIR=/usr/local/bin sh

uv python install 3.14
uv sync --extra dev
source .venv/bin/activate

cd frontend
npm ci
npm run build
cd ..

make
make run

Hope this covers all dependencies…

Thomas