Final Project [ Presentation Photos, Press Coverage ]
|
|
Analysis of Dynamic Shaping in Unaccompanied Bach by Eric Cheng [ website ]
Using the principles of psychoacoustics, loudness curves are extracted from audio recordings of unaccompanied Bach violin sonatas performed by Menuhin, Milstein, and Heifetz. Dynamic trajectories and strategies are compared and contrasted.
|
|
Following a pianist's paces - exploring the information revealed by tempo changes by Ching-Hua Chuan [ website ]
This project develops a program that allows users to visualize the relations between tempo changes, loudness, and musical scores of real performances in audio recordings. Tempo tracking is provided by users through mouse clicks. The amplitude (loudness) of the audio, and the score obtained from MIDI, are shown. It can also be used as a tool for MIDI and audio alignment.
|
|
An Exploration of KTH Rules Application by Siao Jer [ presentation (html) | mini experiment ]
This project is an extension into a study of emotion through music expression by Roberto Bresin. We look at his results and research, and apply that to the ears of non-expert listeners. Eighty participants are asked to classify six pieces, created by Bresin, into six emotional classes: fear, anger, happiness, sadness, tenderness. The results are reported in the presentation.
|
|
Emotion Reading by Shih-Jer Lin and Tsung-Han (Robert) Chiang [ website | presentation (pdf) ]
The Duration Contrast Rule is a music performance rule in the Director Musices system developed by KTH. It refers to the fact that performers might enhance the contrast between long and short note values by playing the short notes shorter and the long notes longer than nominally written in the score. Our project examines whether, and to what extent, the Duration Contrast Rule alone could predict the emotional response to a musical piece.
|
|
Musical Expression Synthesis by Driving in the Tempo-Loudness Space by Jie Liu [ website | presentation (pdf) ]
Langer and Goebl proposed to use tempo-loudness trajectories to analyze musical expression in expressive performances. This project combines Langner's tempo-loudness space with the ESP driving interface. The user drives a car in the tempo-loudness space, thereby controling the tempo and loudness in real-time. The system guides the user where and how fast to drive. Results show that this could be a interactive tool to generate musical expressios.
|
|
Music of Silence - Analysis of Silent Movie Music by Arpi Mardirossian [ website ]
Many plug-in scores were composed to accompany silent movies. These scores were written for specific situations and assigned an emotion. This set of scores provides a unique set of test files with predefined ground truth. I will conduct analysis of the emotional content of the pieces and compare with the ground truth.
|
|
Emotiongram: Visualization of Emotional Content from Audio by Meghen Miles and Merrick Mosst [ website | presentation (pdf) ]
A minimal but effective feature set for emotion detection is combined with music visualization techniques to represent emotional "coloring" in audio at various time scales. We begin to formulate the basis for an exploratory approach to emotion detection, visualization, and tracking.
|
|
Tempo and Loudness Tracking via Frequency and Time Domain Analysis for Polyphonic Music by Erdem Unal [ website ]
We used filter bank analysis to extract musical notes from polyphonic audio. The important thing here is to accurately define the note onsets so that we can analyze the periodicity of each band-pass filter output, and their effect on perceived tempo. We applied autocorrelation methods to estimate the periodicity of the detected note onsets. We also applied short-term energy analysis to track energy changes in each band for intensity analysis.
|
|
Br-E-DM : Brahms Experiment by Director Musices by Leila Vaziri [ website ]
Director Musices is a program which allows the user to add some rules to the music score and have different performances. It contains a set of additive rules that can change the duration, loudness and harmonic and melodic tension and etc. In this experience we will add rules to the Hungarian Dance no. 4 in F-Sharp Major by Johannes Brahms to find out which combinations of rules will gives us pleasant performance compare to original one.
|
|
Clustering of Expressive Music Performance by Haojun (Leo) Wang [ website | presentation (pdf) ]
This project investigates the patterns of expressive music performance by using data mining techniques. Considering music performances as time series, I find clusters of loudness patterns from music segments that are correlated with expression.
|