Christian Steinmetz - flowEQ: Using machine learning and MATLAB to build an intelligent EQ plugin скачать в хорошем качестве

Christian Steinmetz - flowEQ: Using machine learning and MATLAB to build an intelligent EQ plugin

Christian Steinmetz (Music Technology Group at Universitat Pompeu Fabra) “flowEQ: Using machine learning and MATLAB to build an intelligent EQ plugin” The parametric EQ is one of the most powerful forms of the equalizer and requires training and experience for the audio engineer to use it effectively to achieve the desired timbre. flowEQ uses a disentangled variational autoencoder (β-VAE) in order to provide a high level interface to the parametric EQ. By employing the trained decoder network from our model in the real-time plugin (built using MATLAB), the user can quickly search through the relevant configurations of a five band parametric equalizer using one, two, or three powerful controls. This methodology promotes using one’s ears to determine the proper settings instead of using one’s eyes (i.e. observing transfer functions or specific frequency controls). ---- Join the Audio Programmer Community: https://theaudioprogrammer.com/commun... Support the channel:   / theaudioprogrammer   Website: https://www.theaudioprogrammer.com Instagram:   / theaudioprogrammer   Facebook:   / theaudioprogrammer   Twitter:   / audioprogrammer