top of page
  • Crystal D'Esprit

R2D2 Vox Mutatio in Pure Data

R2D2 Vocal mutation patch in pure data


Pure data patchwork for R2D2 vocal mutation and asset saving
Vox Mutatio - Machine Sound

Above you can see the Vox Mutatio pure data patch that emulates a similar output to the machine sounds of R2D2 from Starwars. This patch currently saves the audio from the microphone into an array called vox and separately enables the user to be able to replay the original vocal recording. The patch enables a four second recording this can be altered by increasing mathematically the sample rate 41,000 by the number of seconds that are required inside of the array's properties.

The recording is then sent though the fiddle~ object that through fast fourier transform (FFT) splits the audio into an initial pitch a bang that fires when the transients are starting and stopping, the volume of the overal file as it plays through and a routed 3 way unpacking (this could be increased through the argument chain, yet splits the audio into pitch bandwidths of the frequency content.

The pitch values that do return back to zero are then used as control in frequency and fed to a phasor~ object and an osc~ object providing both a sine wave and sawtooth wave for each of the two elements utilised in this patch.

To enable the storage of the mutated vocal and the recording and storing of the output a writesf~ object is then also being triggered to record at the same time as the audio is being controlled from the first bang.

There is an independent stop function to enable less of the audio to be recorded upon playback or triggering of the vocal mutation.

This is now a full vocal mutation sampler that can be used to manipulate the vocal into machine sound although without abstractions at the moment the patch is already looking very busy and has no user interface that isn't complex so the next steps will be to use abstractions to tidy up the front layer of the vox mutatio synth and then look into further options of how else we can manipulate the recorded audio from the writesf~ file named asset.

This process can be repeated so that further control over the audio inside of pure data can occur in an organised fashion provided the argument names are the same.

The visual arrays are really useful elements to still have on the main user interface (UI) yet it will need to be selective to ensure good organisation.

The patch above will enable a recorded voice to be kept stored, mutated and outputted as an asset. To replay the asset which is exactly what we want to achieve. This will be a good thing when expanding the patch.



bottom of page