Past talks

Speaker: André Jucovsky Bianchi
Date and time: Friday, September 30, 2011 - 16:00
Place: -

Pure Data (Pd) is a realtime signal processing tool widely used for live artistic performances. CUDA is a platform for parallel processing using NVIDIA GPU boards. In this seminar, I will present these technologies and an ongoing work which purpose is to allow for the realtime parallel signal processing using Pd and CUDA.


Speaker: Antonio José Homsi Goulart
Date and time: Tuesday, September 27, 2011 - 10:00
Place: Room 259-A, at IME/USP

The possibility of streaming and the ease to download and store music in computers and portable devices make AMGC (automatic music genre classification) systems a must. Systems based on metadata analysis might be unprecise, and classifications are by artist or by album, rather than by each tune. Other systems, such as those that addopt MFCCs, explores rhytmic and timbric content of the songs, being LDA, SVM and GMM the most used classifiers. The main goal of this work is to investigate if wavelet based entropy, fractal dimension and lacunarity are good parameters to represent music signals and provide a good accuracy in an AMGC system based on these concepts, rather than musical information. The classifiers adopted were SVMs and GMMs. Two databases were created for the tests; one based on 3 genres (blues/classical/lounge) and the other on 4 brazilian genres(axé/bossa-nova/forró/samba).

Speaker: Flávio Luiz Schiavoni
Date and time: Friday, September 2, 2011 - 10:00
Place: -
Speaker: Fabio Kon (IME/USP)
Date and time: Monday, August 29, 2011 - 16:00
Place: Room 256-A, at IME/USP

In this seminar, we will begin by presenting a brief historical overview of works involving interactivity in music production using computational mechanisms. Then, we will present a few academic works developed in Brazil. Finally, we will describe the Mobile Project, which has, as its central theme, the utilization and development of interactive processes in the realm of musical production technologically mediated. The project is divided in multiple lines of research that work in an integrated way: Sonology, Development of Interactive systems, Musical production with Interactive Systems, and Musical acoustics, psychoacoustics and auralization. We will discuss the concepts involved in each of these areas and will present recent artistic works created within the project.

Speaker: Anderson Fraiha Machado (doctorate stutend at IME/USP)
Date and time: Tuesday, August 23, 2011 - 16:00
Place: Room 254-A, at IME/USP

[portuguese only] O problema de Conversão de Voz Inter-Linguística se refere à substituição do timbre ou identidade vocal de uma sentença gravada por um falante (origem) pela de outro falante (destino), supondo que ambos os falantes pronunciam suas respectivas sentenças em línguas diversas. Este problema se difere de uma conversão de voz típica no sentido de que o mapeamento de características acústicas não depende do alinhamento temporal de gravações de sentenças idênticas pronunciadas pelos falantes origem e destino. Será apresentado um arcabouço geral de um sistema de conversão de voz inter-linguística, discutindo o estado-da-arte de algumas das técnicas mais importantes utilizadas em cada passo do processo de conversão de voz.


Subscribe to Upcoming talks Subscribe to Past talks