Skip to content
jackarkydevPublic

About

Vocal processing suite for VCV Rack 2

Resources

Stars

4 stars

Watchers

0 watching

Forks

Latest commit

 

History

28 Commits

Folders and files

Repository files navigation

Astravox

A processing suite for VCV Rack 2: a 16-band channel vocoder for spectral and vocal work, and a three-head tape echo modeled on 1970s tape delay hardware.

Developer: Jack Arky License: GPL-3.0-only


Modules

VOCODER — 42HP

A 16-band channel vocoder with per-band level faders, an internal carrier oscillator, and extensive CV connectivity. Designed as a dedicated vocal processing instrument.

Highlights:

  • 16-band Chamberlin state-variable filter banks (analysis + synthesis)
  • Per-band level faders with live LED meters showing spectral energy
  • Internal VCO: SAW / PULSE with PWM, automatically normalled to carrier input
  • FORMANT control — shifts synthesis bands relative to analysis bands for vowel shaping
  • FREEZE — latches all 16 envelope followers at their current values
  • 16 per-band CV outputs (envelope follower signals, 0–8V)
  • 16 per-band CV inputs with selectable Replace or Sum mode
  • EVEN / ODD mix outputs for stereo spreading
  • Context menu: band spacing (LIN / MEL / BARK), LFO target, Band CV mode

For full documentation, right-click the module in VCV Rack and select Manual.


TAPE ECHO — 20HP

A three-head tape echo and reverb combination modeled on the classic tape-based echo/reverb units of the 1970s. Goes beyond a clean delay: tape hysteresis and saturation, wow & flutter, self-oscillating feedback, and a "Tape Age" wear model reproduce the specific imperfections of a physical tape loop.

Highlights:

  • 3 fixed tape heads (12-position Mode switch selects taps + reverb)
  • Tape hysteresis/saturation feedback loop with measured self-oscillation
  • Convolution-based spring reverb, with an optional reverse-follows-Reverse swell
  • Tape Age wear model — per-instance random defect layout, HF rolloff, wow/flutter, and dropouts above 50% wear
  • Clock sync (patch CLOCK to engage) with note-value divisions and a ±6% Rate nudge trim
  • Gang — links Intensity and the rate axis as one performance gesture
  • Individual H1/H2/H3 head outputs, plus a feedback-loop Send/Return insert point
  • Context menu: Eco mode, machine noise floor, Tape Age presets, Drive tilt, Send/Return routing

For full documentation, right-click the module in VCV Rack and select Manual.


Installation

Download the latest .vcvplugin file from the Releases page.

Requirements: VCV Rack 2.x


Building from Source

Requires the VCV Rack 2 SDK.

git clone https://github.com/JackArkyDev/Astravox.git
cd Astravox
export RACK_DIR=/path/to/Rack-SDK
make
make install    # installs to your Rack user folder

Third-Party Credits

This plugin incorporates the following third-party code and assets:

  • AnalogTapeModel by Jatin Chowdhury — the magnetic hysteresis model used by Tape Echo's tape saturation is adapted from this project. Licensed under GPL-3.0.
  • Inter by The Inter Project Authors — used as a UI typeface. Licensed under the SIL Open Font License 1.1; full license text in res/fonts/OFL.txt.

License

Copyright (C) 2026 Jack Arky

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 3.

See LICENSE for the full license text.

About

Vocal processing suite for VCV Rack 2

Resources

Stars

4 stars

Watchers

0 watching

Forks

Releases

Sponsor this project

Packages

Contributors

Languages