faceWah: Applying wah effect based on face tilt - prototype

Category:ToolTags:
#Music#HCI
Published: 2020 - 6 - 20

"faceWah" is a prototype application that uses facial movements and expressions to control a wah-wah audio effect. Instead of a traditional physical foot pedal, it aims to create an intuitive and expressive interface for sound creation based on facial tilt and mouth opening.

This project is built using a combination of openFrameworks and Max/MSP.

1.
Face
Tracking
(openFrameworks)

Using the C++ framework openFrameworks along with ofxFaceTracker2, the prototype captures a real-time webcam feed to track 68 facial landmarks.

2.
Parameter
Calculation
&
Transmission

From the mapped landmarks, it targets and calculates several geometric values:

  • Nose line tilt (Face angle): Calculated using the top of the nose bridge and the tip of the nose.
  • Mouth opening: The geometric distance between the upper and lower lips.

These extracted values are then transformed and sent across a local network using OSC (Open Sound Control).

3.
Audio
Processing
(Max/MSP)

A Max/MSP patch (wah.maxpat) listens for the incoming OSC data. It maps these facial expressions and tilt numbers to the parameters of a wah effect (such as modulating the cutoff frequency of a filter). As a result, the audio dynamically sweeps and changes based on real-time facial expressions.