Organizational Unit:
School of Music

Research Organization Registry ID
Previous Names
Parent Organization
Parent Organization
Organizational Unit
Includes Organization(s)

Publication Search Results

Now showing 1 - 5 of 5
  • Item
    Computational modeling of improvisation in Turkish folk music using Variable-Length Markov Models
    (Georgia Institute of Technology, 2011-08-31) Senturk, Sertan
    The thesis describes a new database of uzun havas, a non-metered structured improvisation form in Turkish folk music, and a system, which uses Variable-Length Markov Models (VLMMs) to predict the melody in the uzun hava form. The database consists of 77 songs, encompassing 10849 notes, and it is used to train multiple viewpoints, where each event in a musical sequence are represented by parallel descriptors such as Durations and Notes. The thesis also introduces pitch-related viewpoints that are specifically aimed to model the unique melodic properties of makam music. The predictability of the system is quantitatively evaluated by an entropy based scheme. In the experiments, the results from the pitch-related viewpoints mapping 12-tone-scale of Western classical theory and 17 tone-scale of Turkish folk music are compared. It is shown that VLMMs are highly predictive in the note progressions of the transcriptions of uzun havas. This suggests that VLMMs may be applied to makam-based and non-metered musical forms, in addition to Western musical styles. To the best of knowledge, the work presents the first symbolic, machine-readable database and the first application of computational modeling in Turkish folk music.
  • Item
    Towards expressive melodic accompaniment using parametric modeling of continuous musical elements in a multi-attribute prediction suffix trie framework
    (Georgia Institute of Technology, 2010-11-22) Mallikarjuna, Trishul
    Elements of continuous variation such as tremolo, vibrato and portamento enable dimensions of their own in expressive melodic music in styles such as in Indian Classical Music. There is published work on parametrically modeling some of these elements individually, and to apply the modeled parameters to automatically generated musical notes in the context of machine musicianship, using simple rule-based mappings. There have also been many systems developed for generative musical accompaniment using probabilistic models of discrete musical elements such as MIDI notes and durations, many of them inspired by computational research in linguistics. There however doesn't seem to have been a combined approach of parametrically modeling expressive elements in a probabilistic framework. This documents presents a real-time computational framework that uses a multi-attribute trie / n-gram structure to model parameters like frequency, depth and/or lag of the expressive variations such as vibrato and portamento, along with conventionally modeled elements such as musical notes, their durations and metric positions in melodic audio input. This work proposes storing the parameters of expressive elements as metadata in the individual nodes of the traditional trie structure, along with the distribution of their probabilities of occurrence. During automatic generation of music, the expressive parameters as learned in the above training phase are applied to the associated re-synthesized musical notes. The model is aimed at being used to provide automatic melodic accompaniment in a performance scenario. The parametric modeling of the continuous expressive elements in this form is hypothesized to be able to capture deeper temporal relationships among musical elements and thereby is expected to bring about a more expressive and more musical outcome in such a performance than what has been possible using other works of machine musicianship using only static mappings or randomized choice. A system was developed on Max/MSP software platform with this framework, which takes in a pitched audio input such as human singing voice, and produces a pitch track which may be applied to synthesized sound of a continuous timbre. The system was trained and tested with several vocal recordings of North Indian Classical Music, and a subjective evaluation of the resulting audio was made using an anonymous online survey. The results of the survey show the output tracks generated from the system to be as musical and expressive, if not more, than the case where the pitch track generated from the original audio was directly rendered as output, and also show the output with expressive elements to be perceivably more expressive than the version of the output without expressive parameters. The results further suggest that more experimentation may be required to conclude the efficacy of the framework employed in relation to using randomly selected parameter values for the expressive elements. This thesis presents the scope, context, implementation details and results of the work, suggesting future improvements.
  • Item
    Automatic accompaniment of vocal melodies in the context of popular music
    (Georgia Institute of Technology, 2009-04-08) Cao, Xiang
    A piece of popular music is usually defined as a combination of vocal melody and instrumental accompaniment. People often start with the melody part when they are trying to compose or reproduce a piece of popular music. However, creating appropriate instrumental accompaniment part for a melody line can be a difficult task for non-musicians. Automation of accompaniment generation for vocal melodies thus can be very useful for those who are interested in singing for fun. Therefore, a computer software system which is capable of generating harmonic accompaniment for a given vocal melody input has been presented in this thesis. This automatic accompaniment system uses a Hidden Markov Model to assign chord to a given part of melody based on the knowledge learnt from a bank of vocal tracks of popular music. Comparing with other similar systems, our system features a high resolution key estimation algorithm which is helpful to adjust the generated accompaniment to the input vocal. Moreover, we designed a structure analysis subsystem to extract the repetition and structure boundaries from the melody. These boundaries are passed to the chord assignment and style player subsystems in order to generate more dynamic and organized accompaniment. Finally, prototype applications are discussed and the entire system is evaluated.
  • Item
    Generative rhythmic models
    (Georgia Institute of Technology, 2009-04-08) Rae, Alexander
    A system for generative rhythmic modeling is presented. The work aims to explore computational models of creativity, realizing them in a system designed for realtime generation of semi-improvisational music. This is envisioned as an attempt to develop musical intelligence in the context of structured improvisation, and by doing so to enable and encourage new forms of musical control and performance; the systems described in this work, already capable of realtime creation, have been designed with the explicit intention of embedding them in a variety of performance-based systems. A model of qaida, a solo tabla form, is presented, along with the results of an online survey comparing it to a professional tabla player's recording on dimensions of musicality, creativity, and novelty. The qaida model generates a bank of rhythmic variations by reordering subphrases. Selections from this bank are sequenced using a feature-based approach. An experimental extension into modeling layer- and loop-based forms of electronic music is presented, in which the initial modeling approach is generalized. Starting from a seed track, the layer-based model utilizes audio analysis techniques such as blind source separation and onset-based segmentation to generate layers which are shuffled and recombined to generate novel music in a manner analogous to the qaida model.
  • Item
    Hubs and homogeneity: improving content-based music modeling
    (Georgia Institute of Technology, 2008-04-01) Godfrey, Mark Thomas
    With the volume of digital media available today, automatic music recommendation services have proven a useful tool for consumers, allowing them to better discover new and enjoyable music. Typically, this technology is based on collaborative filtering techniques, employing human-generated metadata to base recommendations. Recently, work in content-based recommendation systems have emerged in which the audio signal itself is analyzed for relevant musical information from which models are built that attempt to mimic human similarity judgments. The current state-of-the-art for content-based music recommendation uses a timbre model based on MFCCs calculated on short segments of tracks. These feature vectors are then modeled using GMMs (Gaussian mixture models). GMM modeling of frame-based MFCCs has been shown to perform fairly well on timbre similarity tasks. However, a common problem is that of hubs , in which a relative small number of songs falsely appear similar to many other songs, significantly decreasing the accuracy of similarity recommendations. In this thesis, we explore the origins of hubs in timbre-based modeling and propose several remedies. Specifically, we find that a process of model homogenization, in which certain components of a mixture model are systematically removed, improves performance as measured against several ground-truth similarity metrics. Extending the work of Aucouturier, we introduce several new methods of homogenization. On a subset of the uspop data set, model homogenization improves artist R-precision by a maximum of 3.5% and agreement to user collection co-occurrence data by 7.4%. We also find differences in the effectiveness of the various homogenization methods for hub reduction, with the proposed methods providing the best results. Further, we extend the modeling of frame-based MFCC features by using a kernel density estimation approach to non-parametric modeling. We find that such an approach significantly reduces the number of hubs (by 2.6% of the dataset) while improving agreement to ground-truth by 5% and slightly improving artist R-precision as compared with the standard parametric model. Finally, to test whether these principles hold for all musical data, we introduce an entirely new data set consisting of Indian classical music. We find that our results generalize here as well, suggesting that hubness is a general feature of timbre-based similarity music modeling and that the techniques presented to improve this modeling are effective for diverse types of music.