BACK TO GERHARD-RESEARCH

1 Most Significant Contributions to Research

1.1 Musical and Artistic Interfaces and Interactions

This work includes implementation of various interfaces to audio/musical software and studies of human-computer interactions with various aspects of musical interfaces, including how humans interact with audio editing software programs, audio transcription programs, real-time composition and computer-based orchestral conducting. This work also includes collaboration with working artists and musicians, and development of software tools and methods of interaction which enable this collaboration. Recent publications relating to this work include: [4] which relates to an iPhone/iPad application we created (see Sec. 2.5). Other recent papers relating to music and human interfaces include: [3, 5, 6, 7, 1, 9, 10]

This research is funded primarily by a 2004 CFI grant to establish the Rough Music and Audio Digital Interaction Lab (aRMADILo) as well as NSERC discovery grant support, and local grants from the University of Regina including the SSHRC president’s fund.

1.2 Computational Models of Human Song

My focused individual research area continues to be the development of computational models of the acoustic vocalizations produced when people sing. Research in other areas, for example chord analysis and collaborative creativity, are informing current explorations in this area. Recent publications relating to this work include [2, 17].

This work was also the basis for my PhD Thesis which I note even though it is more than 6 years old because a Technical report (“Pitch extraction and fundamental frequency: history and current techniques”) I wrote as a portion of this thesis has recently become a commonly cited reference for pitch detection, to the point where it has been cited by over 100 papers, and it is the primary reference for the “Pitch Detection Algorithm” article in Wikipedia. This work is funded by NSERC discovery grants

1.3 Interactive Spatialization

This work, with MSc Student JJ Nixdorf, examines issues of human-computer interaction applied to sound spatialization. I have been principally responsible for initiating the project and developing the first version, as well as developing 3-d components of the visual interaction on the latest version. The work is significant because other current spatialization systems are intended for composition of sound spatialization, and cannot be used for real-time interaction [12, 13, 14]. This research is funded primarily by a 2004 CFI grant to establish the Rough Music and Audio Digital Interaction Lab (aRMADILo).

1.4 Collaborative multimedia composition

This work explores the interaction of individuals with collaborators when composing multimedia. Parameter exploration is studied, especially in varying media modes, and systems are developed to allow composers experienced in disparate media to work together on a single collaborative piece. This work is significant because it lays out the underlying parametric structure of such a collaborative composition environment, and has been the basis of the development of structures for both interaction and personalization of multimedia composition. Ongoing work related to this research is published in [15, 16]. This work is funded by NSERC discovery grants

2 Research Contributions and Practical Applications

In the publication venues listed below, contribution is indicated by author order, with the first author typically contributing the most to the work. Unless otherwise indicated as [idea:], I am the originator of the research ideas in all publications. When a student has done significant development or implementation work on an idea, I give them first authorship. Publication venues are typically chosen from among top-rated high-profile music and interaction conferences.

2.1 Articles in refereed publications

[1]
David Gerhard, Brett Park, and Jarrod Ellis (2008). Focus-Plus-Context Audio Interaction Design. Computer Music Modeling and Retrieval, Lecture Notes in Computer Science. 453–477. Citeseer Impact Factor 0.75 (top 34.64%)
[2]
David Gerhard (2005). Multiresolution pitch analysis of talking, singing, and the continuum between. Rough Sets, Fuzzy Sets, Data Mining and Granular Computing), Lecture Notes in Artificial Intelligence (LNAI) 3642, Vol II, pp 294–303. acceptance rate 44%

2.2 Other refereed contributions: Book Chapters

[3]
David Gerhard, Xinglin Zhang (2010) Chord Analysis Using Ensemble Constraints. Advances in Music Information Retrieval. Ras, Zbigniew W. and Wieczorkowska, Alicja (Eds.) ISBN 978-3-642-11673-5.

2.3 Other refereed contributions

[4]
Steven Maupin, David Gerhard, Brett Park (2011). Isomorphic Tessellations for Musical Keyboards. Proceedings of the Sound and Music Computing Conference 2011, Padova, Italy.
[5]
Lijuan Peng and David Gerhard (2009) A Gestural Interface for Orchestral Conducting Education. First International Conference on Computer Supported Education (CSEDU), Lisbon, Portugal.
[6]
Lijuan Peng and David Gerhard (2009) A Wii-based gestural interface for computer conducting systems. Ninth International Conference on New Interfaces for Musical Expression (NIME), Pittsburgh, PA.
[7]
Xinglin Zhang and David Gerhard (2008). Chord Recognition using Instrument Voicing Constraints. International Conference on Music Information Retrieval (ISMIR), Philadelphia, PA, 33–38
[8]
Daryl H. Hepting , L. Peng , T. J. Maciag, David Gerhard and B. Maguire. (2008) Creating synergy between usability courses and open source software projects. ACM SIGCSE Bulletin. ACM Press. 120–123 (Reviewed Professional Magazine article) [idea: Hepting]
[9]
David Gerhard and Ellen Moffat (2007). convocare_consonare: A Duet in Four Voices. International Computer Music Conference, Copenhagen.
[10]
David Gerhard and Jarrod Ellis (2007). Focus-Plus-Context Displays for Audio Interaction. International Conference on Computer Music, Copenhagen.
[11]
Brien Beattie, Garrett Nicolai, David Gerhard, Robert J. Hilderman (2007) Pattern Classification in No-Limit Poker: A Head-Start Evolutionary Approach. Canadian Conference on AI 2007: 204-215 [idea: Beattie/Nicolai]
[12]
JJ Nixdorf and David Gerhard (2006). RITZ: A real-time tool for interactive spatialization. Proc. ACM Multimedia, Santa Barbara, pp 687–690.

acceptance rate 35%, Citeseer Impact Factor 1.22 (top 14.98%)

[13]
JJ Nixdorf and David Gerhard (2006). Real-time sound source spatialization as used in challenging bodies: implementation and performance. International Conference on New Interfaces for Musical Expression (NIME06), Paris, 318–321.

Acceptance rate 44%

[14]
David Gerhard and JJ Nixdorf (2006). Computational sound source localization for musical expression in live performance (Abstract). Spanning the Distance: Canadian meeting of the International Association for the Study of Popular Music. Regina, SK.
[15]
David Gerhard and Daryl H. Hepting (2005). A framework for personalization of interactive sound synthesis. International Computer Music Conference, Barcelona, Spain.

Acceptance rate: 55% [idea: Gerhard/Hepting]

[16]
Daryl H. Hepting, David Gerhard, Joel Rathgaber (2005). Realtime interactive multimedia performance. ACM SIGGRAPH 2005, Los Angeles, California. [idea: Gerhard/Hepting]
[17]
David Gerhard (2005). Pitch track target deviation in natural singing. Sixth International Conference on Music Information Retrieval (ISMIR) Queen Mary, U. of London, pp 514–519.
[18]
Lu Meng and David Gerhard (2005). Acoustic ray tracing for 3D environment simulation. Proc. Canadian Acoustical Assoc Acoustics Week in Canada, London, Ontario. [idea: Meng]

2.4 Non-refereed contributions

[19]
David Gerhard (2006). Review of the 6th international conference on music information retrieval (ISMIR 2005). Computer Music Journal Volume 30, Number 2, 90–92.

2.5 Contributions to practical applications of knowledge

I have a software company that I and two colleagues started. Based on the work published in [4] we have created an iPhone/iPad application (Musix) and made it available for purchase, showing a direct influence of this technology development on the economy of Canada. To date more than ten thousand individual copies of the software have been purchased or downloaded, and more than 400 people use Musix at least once a week.

As listed in Section 3.3 “presentations, shows and lectures”, I routinely work with artists and creative professionals to encorporate technology into their artistic practice, which is a significant practical application of computer technology to the culture of Canada.

3 Other Evidence of Impact and Contributions

3.1 Memberships and Committees

3.2 Consulting Activities and participation in research community

3.3 Presentations, Shows and Lectures

3.4 Public Awareness/Education

4 Delays in Research Activity

I was on full-time Parental Leave from October 1, 2008 to April 8, 2009.

5 Training of Highly Qualified Personnel

I am committed to the ongoing training of HQP, starting from the undergraduate level. Many of the students I have supervised at the MSc level began their studies with me at the Undergraduate level. I also actively participate in interdisciplinary supervision of HQP, with students in cross-disciplinary programs between Computer Science and Engineering, as well as Computer Science and Fine Arts. Successful graduate students from my lab are active members of society, mostly working in gaming and the IT industry. For example, HQP graduating from my supervision now work at Electronic Arts, Talking Dog Studios, Sasktel, and many other high-profile information-technology-based industries.

I usually supervise on average 5 students.

I routinely supervise undergraduate students in the NSERC USRA program, with one or two undergrads under my supervision receiving this award each summer.