Search

The Jam Station is a collaborative musical experience (CME), with embedded sensors to detect, evaluate, and reinforce collaboration among players of all skill levels. The system is also gamified, consisting of four instruments that face a centrally located, vertically oriented, display that visualizes various game states. Playing any of the instruments affect the game as musical collaboration is assessed and visualized on the display. The display also includes a “progress bar’’ which fills up the more players collaborate musically, and to “win’’ the game, the progress bar must be full, initiating a multi-color light show for the players. By creating an environment focused primarily on collaboration, we aim to support collaborative music-making for musicians of all skill levels.

Music improvisation, which is defined as creatively and playfully exchanging musical concepts, is often associated with expert musicianship and musical skill. While there are numerous methods for live scoring improvisation, there are not many resources aimed at assisting novice performance during improvisational activities. Jam Tabs is a notation system that helps coordinate idiomatic musical expression for novice musicians using color-coded instruments and a colored visual display. Jam Tabs utilizes a seven-color notation system, LED cubes and a colored piano keyboard to assist novices in coordinating chord progressions—a common musical element used among improvising musicians in popular western music contexts. We have observed that novice piano players could follow chord progressions in tonal music note for note using color coordination while improvising. We found that seven colors in any key provide enough information for players to experience a satisfactorily expressive and creative jam session.

Sequence-to-sequence models have proven to be highly successful in learning morphological inflection from examples as the series of SIGMORPHON/CoNLL shared tasks have shown. It is usually assumed, however, that a linguist working with inflectional examples could in principle develop a gold standard-level morphological analyzer and generator that would surpass a trained neural network model in accuracy of predictions, but that it may require significant amounts of human labor. In this paper, we discuss an experiment where a group of people with some linguistic training develop 25+ grammars as part of the shared task and weigh the cost/benefit ratio of developing grammars by hand. We also present tools that can help linguists triage difficult complex morphophonological phenomena within a language and hypothesize inflectional class membership. We conclude that a significant development effort by trained linguists to analyze and model morphophonological patterns are required in order to surpass the accuracy of neural models.