This article presents an interface for navigating a musical parameter space. The entire combinatorial space of a ten-parameter synthesizer is laid out as a two-dimensional surface on a multi-touch screen. The surface can be scrolled and zoomed using touch-screen swipe and pinch gestures, reminiscent of a map application. The user can place markers on the surface to flag favorites and to explore regions of different sizes around these points. The mapping from the two-dimensional surface to the higher-dimensional parameter space uses a space-filling curve. Hilbert curves constructed from Gray codes with long bit runs can be used to preserve locality as much as is possible, while still maintaining access to all possibilities. A user study was performed to compare this parameter mapping with a more traditional one-slider-per-parameter interface. Questionnaire responses indicate that different mapping strategies suit different stages of the creative process. The combination of the two interfaces was deemed more useful than either individually, reinforcing the notion that a combination of divergent and convergent processes is important for creative tasks.

In this work, a probabilistic model for multiple-instrument automatic music transcription is proposed. The model extends the shift-invariant probabilistic latent component analysis method, which is used for spectrogram factorization. Proposed extensions support the use of multiple spectral templates per pitch and per instrument source, as well as a time-varying pitch contribution for each source. Thus, this method can effectively be used for multiple-instrument automatic transcription. In addition, the shift-invariant aspect of the method can be exploited for detecting tuning changes and frequency modulations, as well as for visualizing pitch content. For note tracking and smoothing, pitch-wise hidden Markov models are used. For training, pitch templates from eight orchestral instruments were extracted, covering their complete note range. The transcription system was tested on multiple-instrument polyphonic recordings from the RWC database, a Disklavier data set, and the MIREX 2007 multi-F0 data set. Results demonstrate that the proposed method outperforms leading approaches from the transcription literature, using several error metrics.