nicthib / mlauvi

MATLAB GUI for data audiovisualization

mlauvi: An audiovisualization tool for neuroimaging datasets

Introduction

MatLab AUdioVIsualizer (MLAUVI) is a tool that can create beautiful audiovisualizations of neuroimaging (and other) datasets. MLAUVI allows you to adjust a variety of parameters to create an effective audiovisualization, and even allows you to save settings if you want to tweak parameters in the future.

Requirements:

• MATLAB (2017 or later - previous versions have not been tested)
• ffmpeg (required for combined audio/video files)
• Some data (example data is included to help you get yours into the proper format)

Installation

1. Clone or download this repository.
2. Make sure you have ffmpeg installed.
   • For mac/linux, ffmpeg can be easily installed using brew: brew install ffmpeg
   • For windows, you can follow this guide for a straightforward installation.
3. In order to point Matlab to your ffmpeg installation, first navigate to the mlauvi folder and then input:
   • (mac/linux) if installed with brew, run readlink $(which ffmpeg) > ffmpegpath.txt. Otherwise, which ffmpeg > ffmpegpath.txt should work.
   • (windows) where ffmpeg > ffmpegpath.txt
4. Open and run mlauvi.m. Make sure to select "Add to Path" if prompted.
5. You are now ready to create some cool audiovisualizations!

Demo: Creating you first audiovisualization

1. Load a demo dataset by choosing Load > from .mat from the toolbar.
2. Choose a .mat file from mlauvi/demodata/.
3. When prompted Choose Temporal Variable, choose the variable T from the list.
4. Similarly, when prompted Choose Spatial Variable, choose the variable S from the list.
5. Your GUI should now update and look like this:

6. Now that you have the raw data loaded, you can either convert the imported data for audiovisualization by pressing the button `Update H'`, or modify some setting before doing so.

data parameters

This box has options for adjusting:

• Framerate of the outputted video (frames per second)
• Brightness
• % Range of data you want to output.

audio parameters

When converting data to audio, timecourses are mapped onto a sequence of notes, with the first timecourse as the lowest note and each subsequent timecourse a higher note. These options control the parameters of that scale and when a note is played.

• Threshold, the minimum value at which a note is played
• Octave shift, an integer that increases/decreases the pitch of the outputted audio
• Scale type, the arrangement of notes you want
  • As a general rule, the note range should fit on a standard 88 key piano. You will see a warning in the Status box if you selected a note arrangement that does not fit within this range.
  • For data with very few timecourses (~10), you can spread the notes out much more than if you have a larger number of timecourses
  • See the Scale section for more details on what these options mean.
• Key, the note that your first component