This project will automate the process of forming visual shows based on audio processing. Similar to the way computer based visualizers parse audio files, this project will process audio signals and use them to generate a photo visualizer. Based on different properties of music such as beat, tempo, frequency range, and volume, it’s possible to map these properties onto a photo visualizer design. At the conclusion of the project our program will take in an audio file and a series of user chosen photographs and generate a show with morphing and effects based on qualities of the audio file.
This project essentially consists of two main tasks. The first is to break down the audio files themselves. Given an audio file, such as an MP3, the program must be able to interpret the audio file and convert it into data that is easily read. For example, changes in dynamics and intensity must be detected and analyzed. In the second task, the program will create a visual show using photos given by the user. The photos will use functions such as morphing, line detection, and various transition effects to create a captivating yet personalized show. Since the user will associate characteristics to certain photos, such as mood (happy, vibrant, sad, etc.) or even song attributes (temp, tone, etc.) our program will aim to create a show using the photos and transition smoothly between photos based on the characteristic of the song. For instance, if there were instances where a song moved between a “happy” up-tempo section to a “sad” slow section, our visualizer will seek to create the smoothest morphing transition possible between a “happy” photo and a “sad” photo. Currently, there are methods of presenting photos and images of a user in the form of a simple slideshow, but we will create a more interesting, yet still automated way of sharing and exhibiting personal photos.