README for VOGRE - VOice GRains with Emotion Kiran Kumar 17 May, 2016 Demo available at https://www.youtube.com/watch?v=puXWMZ6bUCA Introduction: ================================= VOGRE is a granular synthesis engine that allows for the creation of two simultaneous layers of audio stream. The original source material is a collection of three (3) audio files: - Hi_Happy.wav - Hi_Neutral.wav - Hi_Sad.wav In this current package, six (6) additional audio files are also included: - Hi_Happy_To_Neut.wav - Hi_Happy_To_Sad.wav - Hi_Neut_To_Happy.wav - Hi_Neut_To_Sad.wav - Hi_Sad_To_Happy.wav - Hi_Sad_To_Neut.wav These nine (9) audio files are included in the accompanying Audio_Sources directory and are automatically referenced and used by the code. In addition, you should have the following files, all in the main directory: - detect_pitch_yin.m - front_End.fig - front_End.m - granulate.m - main.m - makeGrain.m - pitchShift.m - preProcessing.m - timeStretch.m - transformPitch.m You should also have an empty Audio_Output directory. Audio that you generate and subsequent write to a file will be automatically saved in this directory. Installation: ================================= No installation should be necessary. This code should run right out of the box. Running the Program: ================================= Just run main.m from the central directory (where the code lives), and you're good to go! If you so choose, you can also execute the preProcessing.m script before running main (or uncomment the appropriate lines in main.m to do it from there). The pre-processing runs the code that loads, pitch detects, and transforms the audio sources into six more pitch-adjusted files. The resulting six files are necessary for the program to work, but they are already included in the appropriate directory for your convenience.