Converting VAD into Ekman

#3
by mirix - opened

I have developed a procedure to translate the VAD values produced by this model into Ekman's basic emotions:

https://github.com/mirix/approaches-to-diarisation/tree/main/emotions

Each emotion paradigm has a different scope of application and I though it would be useful to be able to translate one into another.

This work is loosely inspired by:

https://doi.org/10.3390/electronics10212643

Hi Mirix,

please tell, how did you convert VAD values of the model ([0...1]) to Euclidian space ([-1..1]? Model values seems to be always around 0,3... each.

audmax changed discussion status to closed

Sign up or log in to comment