Matching Items (4)
Filtering by

Clear all filters

Electronic Music Composition and Production

Description
This creative project thesis involves electronic music composition and production, and it uses some elements of algorithmic music composition (through recurrent neural networks). Algorithmic composition techniques are used here as a tool in composing the pieces, but are not the main focus. Thematically, this project explores the analogy between artificial…

This creative project thesis involves electronic music composition and production, and it uses some elements of algorithmic music composition (through recurrent neural networks). Algorithmic composition techniques are used here as a tool in composing the pieces, but are not the main focus. Thematically, this project explores the analogy between artificial neural networks and neural activity in the brain. This project consists of three short pieces, each exploring these concept in different ways.
ContributorsKarpur, Ajay (Author) / Suzuki, Kotoka (Thesis director) / Ingalls, Todd (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
135382-Thumbnail Image.png

Electronic Body Protectors: Improving upon an unbiased method to judge Taekwondo Competitions

Description
In competitive Taekwondo, Electronic Body Protectors (EBPs) are used to register hits made by players during sparring. EBPs are comprised of three main components: chest guard, foot sock, and headgear. This equipment interacts with each other through the use of magnets, electric sensors, transmitters, and a receiver. The receiver is…

In competitive Taekwondo, Electronic Body Protectors (EBPs) are used to register hits made by players during sparring. EBPs are comprised of three main components: chest guard, foot sock, and headgear. This equipment interacts with each other through the use of magnets, electric sensors, transmitters, and a receiver. The receiver is connected to a computer programmed with software to process signals from the transmitter and determine whether or not a competitor scored a point. The current design of EBPs, however, have numerous shortcomings, including sensing false positives, failing to register hits, costing too much, and relying on human judgment. This thesis will thoroughly delineate the operation of the current EBPs used and discuss research performed in order to eliminate these weaknesses.
ContributorsSpell, Valerie Anne (Author) / Kozicki, Michael (Thesis director) / Kitchen, Jennifer (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
147587-Thumbnail Image.png

Data Representation for Predicting Harmonic Clusters with LSTM

Description

The purpose of this project is to create a useful tool for musicians that utilizes the harmonic content of their playing to recommend new, relevant chords to play. This is done by training various Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs) on the lead sheets of 100 different jazz…

The purpose of this project is to create a useful tool for musicians that utilizes the harmonic content of their playing to recommend new, relevant chords to play. This is done by training various Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs) on the lead sheets of 100 different jazz standards. A total of 200 unique datasets were produced and tested, resulting in the prediction of nearly 51 million chords. A note-prediction accuracy of 82.1% and a chord-prediction accuracy of 34.5% were achieved across all datasets. Methods of data representation that were rooted in valid music theory frameworks were found to increase the efficacy of harmonic prediction by up to 6%. Optimal LSTM input sizes were also determined for each method of data representation.
ContributorsRangaswami, Sriram Madhav (Author) / Lalitha, Sankar (Thesis director) / Jayasuriya, Suren (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2021-05
132715-Thumbnail Image.png

Investigation and Analysis of Music Genre Identification via Machine Learning

Description
Modern audio datasets and machine learning software tools have given researchers a deep understanding into Music Information Retrieval (MIR) applications. In this paper, we investigate the accuracy and viability of using a machine learning based approach to perform music genre recognition using the Free Music Archive (FMA) dataset. We…

Modern audio datasets and machine learning software tools have given researchers a deep understanding into Music Information Retrieval (MIR) applications. In this paper, we investigate the accuracy and viability of using a machine learning based approach to perform music genre recognition using the Free Music Archive (FMA) dataset. We compare the classification accuracy of popular machine learning models, implement various tuning techniques including principal components analysis (PCA), as well as provide an analysis of the effect of feature space noise on classification accuracy.
ContributorsKhondoker, Farib (Co-author) / Wildenstein, Diego (Co-author) / Spanias, Andreas (Thesis director) / Ingalls, Todd (Committee member) / Electrical Engineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-05