Robot: Tune Yourself! Automatic Tuning for Musical Robotics

Jim Murphy, Paul Mathews, Ajay Kapur, and Dale Carnegie

Proceedings of the International Conference on New Interfaces for Musical Expression

Abstract

This paper presents a method for a self-tuning procedure for musical robots capable of continuous pitch-shifting. Such a technique is useful for robots consisting of many strings: the ability to self-tune allows for long-term installation without human intervention as well as on-the-fly tuning scheme changes. The presented method consists of comparing a detuned string's pitch at runtime to a pre-compiled table of string responses at varying tensions. The behavior of the current detuned string is interpolated from the two nearest pre-characterized neighbors, and the desired virtual fret positions are added to the interpolated model. This method allows for rapid tuning at runtime, requiring only a single string actuation to determine the pitch. After a detailed description of the self-tuning technique and implementation, the results will be evaluated on the new Swivel 2 robotic slide guitar. The paper concludes with a discussion of performance applications and ideas for subsequent work on self-tuning musical robotic systems.

Citation

Jim Murphy, Paul Mathews, Ajay Kapur, and Dale Carnegie. 2014. Robot: Tune Yourself! Automatic Tuning for Musical Robotics. Proceedings of the International Conference on New Interfaces for Musical Expression. DOI: 10.5281/zenodo.1178883

BibTeX Entry

@inproceedings{jmurphy2014,
 abstract = {This paper presents a method for a self-tuning procedure for musical robots capable of continuous pitch-shifting. Such a technique is useful for robots consisting of many strings: the ability to self-tune allows for long-term installation without human intervention as well as on-the-fly tuning scheme changes. The presented method consists of comparing a detuned string's pitch at runtime to a pre-compiled table of string responses at varying tensions. The behavior of the current detuned string is interpolated from the two nearest pre-characterized neighbors, and the desired virtual fret positions are added to the interpolated model. This method allows for rapid tuning at runtime, requiring only a single string actuation to determine the pitch. After a detailed description of the self-tuning technique and implementation, the results will be evaluated on the new Swivel 2 robotic slide guitar. The paper concludes with a discussion of performance applications and ideas for subsequent work on self-tuning musical robotic systems.},
 address = {London, United Kingdom},
 author = {Jim Murphy and Paul Mathews and Ajay Kapur and Dale Carnegie},
 booktitle = {Proceedings of the International Conference on New Interfaces for Musical Expression},
 doi = {10.5281/zenodo.1178883},
 issn = {2220-4806},
 month = {June},
 pages = {565--568},
 publisher = {Goldsmiths, University of London},
 title = {Robot: Tune Yourself! Automatic Tuning for Musical Robotics},
 url = {http://www.nime.org/proceedings/2014/nime2014_345.pdf},
 year = {2014}
}