In this thesis we exploit piano acoustics to automatically transcribe piano recordings into a symbolic representation: the pitch and timing of each detected note. To do so we use approaches based on non-negative matrix factorisation (NMF). To motivate the main contributions of this thesis, we provide two preparatory studies: a study of using a deterministic annealing EM algorithm in a matrix factorisation-based system, and a study of decay patterns of partials in real-word piano tones. Based on these studies, we propose two generative NMF-based models which explicitly model different piano acoustical features. The first is an attack/decay model, that takes into account the time-varying timbre and decaying energy of piano sounds. The system divides a piano note into percussive attack and harmonic decay stages, and separately models the two parts using two sets of templates and amplitude envelopes. The two parts are coupled by the note activations. We simplify the decay envelope by an exponentially decaying function. The proposed method improves the performance of supervised piano transcription. The second model aims at using the spectral width of partials as an independent indicator of the duration of piano notes. Each partial is represented by a Gaussian function, with the spectral width indicated by the standard deviation. The spectral width is large in the attack part, but gradually decreases to a stable value and remains constant in the decay part. The model provides a new aspect to understand the time-varying timbre of piano notes, but furtherinvestigation is needed to use it effectively to improve piano transcription. We demonstrate the utility of the proposed systems in piano music transcription and analysis. Results show that explicitly modelling piano acoustical features, especially temporal features, can improve the transcription performance.

The purpose of this treatise is to discuss tone production on the piano, specifically focusing on the research of Otto Rudolph Ortmann, whose work marked a turning point in the history of piano pedagogy and set a new standard for piano-related scholarship. Writing in the early twentieth century, Ortmann, who was both an accomplished pianist and an avid scientist, was one of the first to consciously and meticulously combine the two fields. Today, Ortmann's books are mostly out of print and his research is little known by the average piano student or teacher. However, Ortmann's books greatly affected and influenced many of his contemporaries, and no matter how neglected today, they contain a wealth of information and practical advice highly relevant to any serious pianist. While his research spans a range of topics, including the science of piano acoustics, experiments and scientific explanations of piano technique, and even music education, his work focusing on piano tone production was perhaps the most controversial. In this treatise, Ortmann's concepts, experiments, and conclusions related to tone production will be discussed. Certain scientific elements related to tone production, such as the acoustics of sound, the mechanism of the piano, and the relevant aspects of basic physics, will serve as an introduction to this discussion. A historical overview of the pedagogical trends surrounding the subject will show how Ortmann's work has been sometimes overlooked and often misinterpreted. A thorough analysis of Ortmann's research will demonstrate its balanced approach and its indispensable relevance to the modern pianist and pedagogue. Finally, his work will be used as a vantage point to shed light on subsequent trends of piano pedagogy and to ask questions about the role of tone production in modern pianism. / text