Vita is a Python module for interacting with the Vital Synthesizer. It is not an official product related to Vital. Vita uses Effort-based versioning.
Vita is supported on Linux, macOS, and Windows. Install with pip:
pip install vitafrom scipy.io import wavfile
import vita
sample_rate = 44100
bpm = 120.0
note_dur = 1.0
render_dur = 3.0
pitch = 36 # integer
velocity = 0.7 # [0.0 to 1.0]
synth = vita.Synth()
# The initial preset is loaded by default.
synth.set_sample_rate(sample_rate)
synth.set_bpm(bpm)
# Let's make a custom modulation using
# the available modulation sources and destinations.
# These lists are constant.
print("potential sources:", vita.get_modulation_sources())
print("potential destinations:", vita.get_modulation_destinations())
# "lfo_1" is a potential source,
# and "filter_1_cutoff" is a potential destination.
assert synth.connect_modulation("lfo_1", "filter_1_cutoff")
controls = synth.get_controls()
controls["modulation_1_amount"].set(1.0)
controls["filter_1_on"].set(1.0)
val = controls["filter_1_on"].value()
controls["lfo_1_tempo"].set(vita.constants.SyncedFrequency.k1_16)
# Use normalized parameter control (0-1 range, VST-style)
controls["filter_1_cutoff"].set_normalized(0.5) # Set knob to 50%
print(controls["filter_1_cutoff"].get_normalized()) # Get normalized value
# Get parameter details and display text
info = synth.get_control_details("delay_style")
print(f"Options: {info.options}") # ["Mono", "Stereo", "Ping Pong", "Mid Ping Pong"]
print(f"Current: {synth.get_control_text('delay_style')}") # e.g., "Stereo"
# Render audio to numpy array shaped (2, NUM_SAMPLES)
audio = synth.render(pitch, velocity, note_dur, render_dur)
wavfile.write("generated_preset.wav", sample_rate, audio.T)
# Dump current state to JSON text
preset_path = "generated_preset.vital"
json_text = synth.to_json()
with open(preset_path, "w") as f:
f.write(json_text)
# Load JSON text
with open(preset_path, "r") as f:
json_text = f.read()
assert synth.load_json(json_text)
# Or load directly from file
assert synth.load_preset(preset_path)
# Load the initial preset, which also clears modulations
synth.load_init_preset()
# Or just clear modulations.
synth.clear_modulations()Vita releases the Python GIL during the DSP-heavy work of render, render_file,
load_preset, load_json, and to_json. This means you can render many presets
in true parallel with a ThreadPoolExecutor instead of multiprocessing -- no
process-spawn or pickling overhead, shared memory, and (importantly) no fork(),
so it works even after importing thread-spawning libraries like JAX where a
fork()-based pool would crash.
Give each thread its own vita.Synth. Sharing one Synth across threads is
safe -- its critical section serializes concurrent calls -- but you get no
parallelism from it, since only one thread renders at a time. See
examples/multithreading_presets.
from concurrent.futures import ThreadPoolExecutor
import threading, vita
_local = threading.local()
def render_preset(path):
synth = getattr(_local, "synth", None)
if synth is None:
synth = _local.synth = vita.Synth() # one Synth per worker thread
synth.load_preset(path)
return synth.render(60, 0.7, 1.0, 2.0)
with ThreadPoolExecutor(max_workers=8) as pool:
audios = list(pool.map(render_preset, preset_paths))setup.py compiles the extension, so an ordinary pip install is enough:
git clone --recursive https://github.com/DBraun/Vita.git
cd Vita
pip install .You need a C++17 toolchain and CMake. The compile itself is driven by the
Projucer-generated project for your platform -- make on Linux, xcodebuild on
macOS, msbuild on Windows -- which the build invokes for you against the Python
interpreter running the install. Type stubs (vita.pyi) are generated as part of
the build.
Full documentation is at dbraun.github.io/Vita, including the API reference and the changelog. To build the docs locally:
pip install -r docs/requirements.txt
sphinx-build -b html docs docs/_build/htmlIf you find any issues with the code, report them at https://github.com/DBraun/Vita.
If you are making a proprietary or closed source app and would like to use Vital's source code, contact licensing@vital.audio for non GPLv3 licensing options.
The source code is licensed under the GPLv3. If you download the source or create builds you must comply with that license.
- Do not create an app and distribute it on the iOS app store. The app store is not comptabile with GPLv3 and you'll only get an exception for this if you're paying for a GPLv3 exception for Vital's source (see Code Licensing above).
- Do not use the name "Vital", "Vital Audio", "Tytel" or "Matt Tytel" for marketing or to name any distribution of binaries built with this source. This source code does not give you rights to infringe on trademarks.
- Do not connect to any web service at https://vital.audio, https://account.vital.audio or https://store.vital.audio from your own builds. This is against the terms of using those sites.
- Do not distribute the presets that come with the free version of Vital. They're under a separate license that does not allow redistribution.