Carnegie Mellon Computer Music Group

Research Seminars & Other Events

We meet approximately once every two-three weeks during the Fall and Spring semesters to discuss the latest in computer music and sound synthesis. EMAIL LIST: If you would like to be added to our email list to be informed about future presentations, please send email to Tom Cortina (username: tcortina, domain: cs.cmu.edu).

For other semesters, click here:
FALL 2013 | SPRING 2013| FALL 2012 | SPRING 2009 | FALL 2008 | SPRING 2008 | FALL 2007 | SPRING 2007 | FALL 2006 | SPRING 2006 | FALL 2005 | SUMMER 2005 | SPRING 2005

FALL 2008

RESEARCH SEMINARS

Thursday, December 4 -- 1:00-2:00PM Wean 3213 (Computer Music Lab) - NOTE: ROOM CHANGE
Listening and Discussion: New Computer Music

Join us in Roger's lab for listening and discussion of new computer music including work from ICMC 2008. Feel free to bring new computer music that you're interested in.

Thursday, November 20 -- 1:00-2:00PM Newell Simon 3001
Topic: Predicting the Next Beat in a Song
Speaker: Nathaniel Anozie, Graduate Student, Department of Statistics, Carnegie Mellon University

Abstract: We predict the location at which beats occur in music using foot taps and hand taps. The main contribution of this work is two-fold: first we compare Local Smoothing methods to a particular Kalman Filter method, and second we estimate theoretical confidence intervals for prediction error.

SPECIAL DATE AND TIME
Tuesday, November 11 -- 2:30-3:30PM Newell Simon 3305

Computational Thinking Seminar Series
Speaker: Noel Zahler, Head, School of Music, Carnegie Mellon University

Abstract: The use of computational thinking in music has become ubiquitous, even though many end users don't realize it. To this end, I will give a brief overview of the most essential areas where such thinking is applied and explain some of the speculative uses of these technologies. I will then turn to my own work in building compositional tools and synthetic performers to explain how the formalization of musical thought for composers and the analysis of performance practice have influenced my work as a composer and the manner in which these abstract concepts can influence the teaching of music. It is through the power of abstraction and the metaphors we use to describe process that we realize the potential of the human/machine interface.

Thursday, October 23 -- 1:00-2:00PM Newell Simon 3001
Topic: A Relevance Feedback and Content-based Music Retrieval System
Speaker: Seungmin Rho, Visiting Scholar, Carnegie Mellon University

Abstract: Query-by-humming has been considered as one of the most intuitive and effective query methods for music retrieval. For voice humming to be a reliable query source, elaborate signal processing and acoustic similarity measurement schemes are necessary. In this talk, we will describe a novel music retrieval system called MUSEMBLE (MUSic enEMBLE) based on three distinct features: (i) A sung or hummed query is automatically transcribed into a sequence of pitch and duration pairs with improved accuracy for music representation. More specifically, we developed two new and unique techniques called WAE (Windowed Average Energy) and Dynamic ADF (Amplitude-based Difference Function) onsets for more accurate note segmentation and onset/offset detection in an acoustic signal, respectively. The former improves energy-based approaches such as AE (Average Energy) by defining small but coherent windows with local and global threshold values. On the other hand, the latter improves the AF (Amplitude Function) that calculates the summation of the absolute values of signal differences for the clustering energy contour. (ii) For the indexing purpose, we proposed a popularity-adaptive indexing structure called FAI (Frequently Accessed Index) based on frequently queried tunes. (iii) A user query is reformulated using user relevance feedback with a genetic algorithm to improve retrieval performance. Even though we have focused on humming queries in this work, MUSEMBLE provides versatile query and browsing interfaces for various kinds of users. We have carried out extensive experiments on the prototype system to evaluate the performance of our voice query transcription and genetic algorithm-based relevance feedback (RF) schemes. We demonstrate that our proposed method improves the retrieval accuracy up to 20-40% compared with other popular RF methods.

Thursday, October 9 -- CANCELLED

Thursday, September 25 -- 1:00-2:00PM Newell Simon 3001
Topic: ISMIR 2008
Speaker: Roger Dannenberg

Abstract: Roger will present some of the highlights of the International Conference on Music Information Retrieval (ISMIR) 2008 held at Drexel University in mid September.

Thursday, September 11 -- 1:00-2:00PM Newell Simon 3001
Topic: ICMC 2008
Speaker: Roger Dannenberg

Abstract: Roger will present some of the new research highlighted at the International Computer Music Conference (ICMC) 2008 being held in Belfast from August 24-29.

FUTURE SEMINAR DATES

If you would like to present a topic at our seminar, please send email to Tom Cortina (username: tcortina, domain: cs.cmu.edu).

Web page and seminar program managed by Tom Cortina, CSD