The Research/Technology track is designed for students interested in musically-related research or developing new music technology. Students in this track study with a departmental advisor, and if applicable to a specific research topic, with another member of the broader Johns Hopkins academic community. Research/Technology track students work closely with practicing composer and performers in developing new computer music systems. Areas of research may include psychoacoustics, perception, hardware or software synthesis and control techniques, algorithmic composition and performance systems, and other related topics.
Gradus ad Parnassum (Species Counterpoint Software)
Developed in collaboration with Johns Hopkins Center for Educational Research and the Peabody Music Theory department, Gradus ad Parnassum is a java webstart application that allows first year music theory students to practice Voice Leading Excercises from the comfort of their own computer.
The project is currently under development by Patrick Donnelly and will be released at the CER's annual symposium in April 2007.
The Hall-O-Deck (Virtual Concert Hall) Project
Utilizing hardware provided by the Intel grant, The Computer Music Department is developing a system with which a performer can rehearse in a practice room, and his instrument's acoustic signal will be processed by NT workstations in such a way that the acoustics of a specific concert hall will be replicated. This is accomplished by the performer playing into a microphone, applying digital signal processing to the signal, and playing back signals into the practice room (through small multimedia speakers) which replicate the acoustics of a desired hall. Initial experiments with "off-the-shelf" processing equipment have been promising. We will, however, need to create models for specific concert halls and develop our own signal processing algorithms.
Intel Grant Projects
The Computer Music Department recently received a grant from the Intel Corporation to pursue several research projects including: The PARIS/Ensoniq/Intelligent Devices Project; Creating a realtime, interactive, networked composition/performance system; and Developing the Virtual Concert Hall.
Gamera: Optical music recognition in a new shell
Researcher: Karl MacMillan, Michael Droettboom, and Ichiro Fujinaga
Target: ICMC2002 (paper PDF)
Optical music interpretation
Researcher: Michael Droettboom, Ichiro Fujinaga, and Karl MacMillan (paper PDF)
Recognition of Isolated Instrument Tones by Conservatory Musicians
Researcher: Asha Srinivasan, David Sullivan, and Ichiro Fujinaga
Target: ICMPC2002 (paper PDF)
Converting Levy Sheet Music Collection to MIDI
Using Optical Music Recognition system, convert the collection of sheet
music to computer-readable format including MIDI, which then can be heard
via web clients.
Optical Music Recognition System within a Large-Scale Digitization Project
Researchers: G. Sayeed Choudhury, Tim DiLauro, Michael Droettboom, Ichiro Fujinaga, Brian Harrington, Karl MacMillan
Target: ISMIR2000 (paper PDF)
Interpreting the semantics of music notation using an extensible and object-oriented system
Researchers: Michael Droettboom, Ichiro Fujinaga
(paper PDF)
Latency of audio and MIDI data over LANs
Measuring latencies of audio and MIDI data over a typical music school
LAN using 10Mb switch, 100Mb hub and 100Mb switch.
Researchers: Tony Willert, Ichiro Fujinaga
Target: ICMC
Timbral recognition using lazy learning
How do we recognize timbre? How well can we recognize using just the
steady-state portion of musical timbre? Can machines do this in
realtime?
Researcher: Ichiro Fujinaga
Machine recognition of timbre using steady-state tone of acoustic
musical instruments
Target: ICMC98 (abstract)
(paper
PDF)
The effect of vibrato on response time in determining the pitch relationship of violin tones
Can the vibrato hide bad intonation?
Researchers: Lilit Yoo, Stephan Moore, David Sullivan, Ichiro Fujinaga
Target: ICMPC98 (abstract)
(paper PDF)
Implementation of exemplar-based learning model for music cognition
The exemplar-based learning model is proposed as an alternative approach to modeling many aspects of music cognition.
Researcher: Ichiro Fujinaga, Stephan Moore, David S. Sullivan, Jr.
Target: ICMPC98 (abstract)
(paper PDF)
Zeta violin latencies and techniques
Zeta violin converts analog sound to MIDI via multi-channel IVLPitchRider hardware.
The problem is the delay in calculating the pitch. The purpose
of this study is to measure the latency which depends on many factors such
as pitch, articulation, dynamics, etc. Also new violin techniques will
be investigated to minimize the latency.
Researchers: Lilit Yoo, Ichiro Fujinaga, Geoffrey Wright
Targets: SEAMUS98 (abstract)
paper read at SEAMUS98, April 18, 1998 (paper PDF)
Real-time software synthesis for psychoacoustic experiments
Introduction to using real-time software synthesis, Super Collider,
MPS, and Pd as valuable tools for psychoacoustic experiments. Small
example experiments will be conducted.
Exemplar-based learning in adaptive optical music recognition system
The learning process of an adaptive optical music recognition system (AOMR) is described here. By
combining k-nearest neighbor classifier and genetic algorithm, the system can learn to recognize
new music symbols and handwritten music notations, and it also continually improves the accuracy
in recognizing these objects. Given the wide range of music notation styles, these are essential
characteristics of a music recognizer.
The Research/Technology track is designed for students interested in musically-related research or developing new music technology. Students in this track study with a departmental advisor, and if applicable to a specific research topic, with another member of the broader Johns Hopkins academic community. Research/Technology track students work closely with practicing composer and performers in developing new computer music systems. Areas of research may include psychoacoustics, perception, hardware or software synthesis and control techniques, algorithmic composition and performance systems, and other related topics.
Gradus ad Parnassum (Species Counterpoint Software)
Developed in collaboration with Johns Hopkins Center for Educational Research and the Peabody Music Theory department, Gradus ad Parnassum is a java webstart application that allows first year music theory students to practice Voice Leading Excercises from the comfort of their own computer.
The project is currently under development by Patrick Donnelly and will be released at the CER's annual symposium in April 2007.
The Hall-O-Deck (Virtual Concert Hall) Project
Utilizing hardware provided by the Intel grant, The Computer Music Department is developing a system with which a performer can rehearse in a practice room, and his instrument's acoustic signal will be processed by NT workstations in such a way that the acoustics of a specific concert hall will be replicated. This is accomplished by the performer playing into a microphone, applying digital signal processing to the signal, and playing back signals into the practice room (through small multimedia speakers) which replicate the acoustics of a desired hall. Initial experiments with "off-the-shelf" processing equipment have been promising. We will, however, need to create models for specific concert halls and develop our own signal processing algorithms.
Intel Grant Projects
The Computer Music Department recently received a grant from the Intel Corporation to pursue several research projects including: The PARIS/Ensoniq/Intelligent Devices Project; Creating a realtime, interactive, networked composition/performance system; and Developing the Virtual Concert Hall.
Gamera: Optical music recognition in a new shell
Researcher: Karl MacMillan, Michael Droettboom, and Ichiro Fujinaga
Target: ICMC2002 (paper PDF)
Optical music interpretation
Researcher: Michael Droettboom, Ichiro Fujinaga, and Karl MacMillan
(paper PDF)
Recognition of Isolated Instrument Tones by Conservatory Musicians
Researcher: Asha Srinivasan, David Sullivan, and Ichiro Fujinaga
Target: ICMPC2002 (paper PDF)
Converting Levy Sheet Music Collection to MIDI
Using Optical Music Recognition system, convert the collection of sheet music to computer-readable format including MIDI, which then can be heard via web clients.
Researcher: Ichiro Fujinaga
NSF grant (1999-2001) abstract, part of Digital Libraries Initiative - Phase 2
Strike Up the Score
Deriving Searchable and Playable Digital Formats from Sheet Music
Researchers: G. Sayeed Choudhury, Tim DiLauro, Michael Droettboom,
Ichiro Fujinaga, Karl MacMillan
D-Lib Magazine, (Feb 2001), Volume 7 Number 2
Using the Web for Live Interactive Music
Researchers: John Young
Networked music: bridging real and virtual space
Researchers: John Young
Organised Sound (2001), 6: 107-110 Cambridge University Press
Optical Music Recognition System within a Large-Scale Digitization Project
Researchers: G. Sayeed Choudhury, Tim DiLauro, Michael Droettboom,
Ichiro Fujinaga, Brian Harrington, Karl MacMillan
Target: ISMIR2000 (paper PDF)
Interpreting the semantics of music notation using an extensible and object-oriented system
Researchers: Michael Droettboom, Ichiro Fujinaga
(paper PDF)
Comparison of pitch detection algorithms
Researchers: Lilit Yoo, Ichiro Fujinaga
Target: ICMC99 (abstract) (paper PDF)
Piano Master Classes via the Internet
Researchers: John Young, Ichiro Fujinaga
Target: ICMC99 (abstract) (paper PDF)
Multimodal input capacities
Researchers: Dave Sullivan, Ichiro Fujinaga
Target: ICMC99 (abstract)
Latency of audio and MIDI data over LANs
Measuring latencies of audio and MIDI data over a typical music school LAN using 10Mb switch, 100Mb hub and 100Mb switch.
Researchers: Tony Willert, Ichiro Fujinaga
Target: ICMC
Timbral recognition using lazy learning
How do we recognize timbre? How well can we recognize using just the steady-state portion of musical timbre? Can machines do this in realtime?
Researcher: Ichiro Fujinaga
Machine recognition of timbre using steady-state tone of acoustic musical instruments
Target: ICMC98 (abstract) (paper PDF)
Toward realtime recognition of acoustic musical instruments
Target: ICMC99 (abstract) (paper PDF)
Violin vibrato technique and its implication for pitch perception
Do violin players slightly flat when playing vibrato?
Researchers: Lilit Yoo, Ichiro Fujinaga
Target: ICMC98 (abstract)
The effect of vibrato on response time in determining the pitch relationship of violin tones
Can the vibrato hide bad intonation?
Researchers: Lilit Yoo, Stephan Moore, David Sullivan, Ichiro Fujinaga
Target: ICMPC98 (abstract) (paper PDF)
Implementation of exemplar-based learning model for music cognition
The exemplar-based learning model is proposed as an alternative approach to modeling many aspects of music cognition.
Researcher: Ichiro Fujinaga, Stephan Moore, David S. Sullivan, Jr.
Target: ICMPC98 (abstract) (paper PDF)
Zeta violin latencies and techniques
Zeta violin converts analog sound to MIDI via multi-channel IVLPitchRider hardware. The problem is the delay in calculating the pitch. The purpose of this study is to measure the latency which depends on many factors such as pitch, articulation, dynamics, etc. Also new violin techniques will be investigated to minimize the latency.
Researchers: Lilit Yoo, Ichiro Fujinaga, Geoffrey Wright
Targets: SEAMUS98 (abstract) paper read at SEAMUS98, April 18, 1998 (paper PDF)
Real-time software synthesis for psychoacoustic experiments
Introduction to using real-time software synthesis, Super Collider, MPS, and Pd as valuable tools for psychoacoustic experiments. Small example experiments will be conducted.
Researchers: Stephan Moore, David Sullivan, Ichiro Fujinaga
Targets: ICMPC98 ( abstract) (notes) (paper HTML) (paper PDF)
Exemplar-based learning in adaptive optical music recognition system
The learning process of an adaptive optical music recognition system (AOMR) is described here. By combining k-nearest neighbor classifier and genetic algorithm, the system can learn to recognize new music symbols and handwritten music notations, and it also continually improves the accuracy in recognizing these objects. Given the wide range of music notation styles, these are essential characteristics of a music recognizer.
Researchers: Ichiro Fujinaga
Target: ICMC96 (paper PDF)