Norse2DynamicFilter: Augmented Vocality Composing Tools #2

As part of the Augmented Vocality project, we have been developing various tools to support Edmund Hunt’s compositonal workflow. Norse2DynamicFilter is the second of these tools, making use of Mohamed El Idrissi’s Praat analyses of Norse words in the Augmented Vocality sample archive.

The patch, designed in SuperCollider, has been designed to allow Edmund to explore ways that the frequency ‘fingerprints’ of words within the archive can be imposed on other sounds. The aim was to create a tool that could audition a soundfile with the word-formant transformation applied, as a way to explore how such transformations might become part of the electronic textures in Edmunds upcoming works for the Augmented Vocality project.

Mohamed’s Praat analyses of Norse words produced a large CSV file, with data describing formant frequencies and strengths within each phoneme of each word. The patch takes a string of these words, as spelled in the archive, and a filepath to a soundfile as input. SuperCollider then fetches the data corresponding to each word and strings them together as a playable sequence of frequency values. The word-frequencies then get ‘played’ into the linked soundfile through resonant bandbass filters. The patch plays back the resulting file on execution, and also saves a copy of the output to SuperColliders default record path.

If you would like to try the patch out for yourself, the file – along with the csv data – is available on GitHub here:

https://github.com/JoeWrightMusic/Norse2DynamicFilter