Live audio synthesis with Synchrotron
summary
Live synthesiser called Synchrotron runs entirely in pure Python using numpy or plain lists to render audio buffers at standard 44.1 kHz, demonstrates a working sine node and a simple FM patch that modulates a 440 Hz carrier, uses a 128-sample buffer to trade latency for stability, accepts MIDI from a Teenage Engineering OP‑1 Field, and streams to playback. It proves real‑time audio is feasible in Python but hits CPU limits and buffer underruns produce audible glitches.
AI-written summary · about
chapters (14)
- 0:00Introduction to Synchrotron
- 1:33Stream processing basics
- 5:02Real-time audio systems
- 6:47Analog vs digital audio
- 9:10Digital audio conversion
- 12:23Sample rate and latency
- 13:05Buffering for stability
- 14:29Building a sine wave node
- 18:47Live demo of Synchrotron
- 21:22Frequency modulation example
- 22:56MIDI keyboard integration
- 24:19Python vs C++ for audio
- 25:22Live coding advantages
- 25:49Q&A session
related talks
29mGenerate buzz with realtime FM audio synthesis
31mSounds good!
20mSoftware-defined declarative synthesizer live-coding in a jupyter notebook
03mBuild a Synthesizer with Python
33mSounds, Synths, and Sonic Pi! Oh My!
37mSonic PI: live music, live coding
37mHow to code music?
47mProgramming In Time - Live Coding for Creative Performances