teslasynth 0.6.4

Python bindings and analysis tools for the teslasynth firmware

teslasynth Python library

Python bindings and offline analysis tools for the Teslasynth MIDI synthesizer firmware.

The library exposes the same synthesis engine used by the ESP32 firmware so you can render MIDI files, visualise pulse signals, and tune safety parameters before connecting any high-voltage hardware.

Licensed under LGPL v3.

Installation

pip install teslasynth

Optional extras:

pip install "teslasynth[plot]"    # Plotly visualisations
pip install "teslasynth[wav]"     # WAV/FLAC export (requires soundfile)
pip install "teslasynth[listen]"  # Live MIDI input -> audio (requires sounddevice + python-rtmidi)

Quick start

from teslasynth import Teslasynth
from teslasynth import wav

synth = Teslasynth()

# Render a MIDI file to a single-channel FLAC (recommended — much smaller than WAV)
wav.write("song.mid", "song.flac", synth=synth)

# Render specific channels into a multichannel file
wav.write("song.mid", "song.flac", synth=synth, channels=[0, 1, 2])

# Render all 8 output channels
wav.write("song.mid", "all.flac", synth=synth, channels=list(range(8)))

CLI

teslasynth render   <midi> <out>    [--config FILE] [--sample-rate HZ] [--channel CHANNELS]
teslasynth listen                   [--config FILE] [--midi-port NAME] [--channel CHANNELS]
                                    [--sample-rate HZ] [--blocksize N] [--audio-device NAME|ID]
                                    [--list-ports]
teslasynth plot     <midi>          [--config FILE] [--out FILE.html] [--channel N]
teslasynth signal   <midi>          [--config FILE] [--out FILE.html] [--channel N]
teslasynth config   [--config FILE] [key=value ...]
teslasynth instruments
teslasynth percussions
teslasynth envelope <instrument>    [--out FILE.html]
teslasynth version

The output format for render is selected from the file extension (.wav or .flac). FLAC is recommended for multichannel exports or long recordings.

# Single channel FLAC (default channel 0)
teslasynth render song.mid out.flac

# Specific channels into a multichannel file
teslasynth render song.mid out.flac --channel 0,1,2

# Range of channels
teslasynth render song.mid out.flac --channel 0-3

# All 8 channels
teslasynth render song.mid out.flac --channel all

Live MIDI playback

listen opens a virtual MIDI port named teslasynth that any DAW or sequencer can connect to, synthesises incoming notes in real time, and streams the result to your audio output. Requires teslasynth[listen].

# Open virtual port "teslasynth", stream channel 0 to default audio device
teslasynth listen

# Stream all 8 output channels (audio device must support 8 outputs)
teslasynth listen --channel all

# Connect to a specific physical MIDI device instead of creating a virtual port
teslasynth listen --midi-port "My Keyboard"

# List available MIDI ports and audio devices
teslasynth listen --list-ports

# Lower latency (256-frame buffer ~= 5.3 ms at 48 kHz)
teslasynth listen --blocksize 256

Documentation

Full documentation at https://teslasynth.hnaderi.dev/python