An augmented lagrangian method for piano transcription using equal loudness thresholding and lstm-based decoding

Abstract

A central goal in automatic music transcription is to detect individual note events in music recordings. An important variant is instrument-dependent music transcription where methods can use calibration data for the instruments in use. However, despite the additional information, results rarely exceed an f-measure of 80%. As a potential explanation, the transcription problem can be shown to be badly conditioned and thus relies on appropriate regularization. A recently proposed method employs a mixture of simple, convex regularizers (to stabilize the parameter estimation process) and more complex terms (to encourage more meaningful structure). In this paper, we present two extensions to this method. First, we integrate a computational loudness model to better differentiate real from spurious note detections. Second, we employ (Bidirectional) Long Short Term Memory networks to re-weight the likelihood of detected note constellations. Despite their simplicity, our two extensions lead to a drop of about 35% in note error rate compared to the state-of-the-art.

A Lightweight Instrument-Agnostic Model for Polyphonic Note Transcription and Multipitch Estimation

“…Multiple methods have been proposed for estimating notes from pitch posteriorgrams e.g. using median filtering [11], Hidden Markov Models [16] or neural networks [20,21]. While most approaches consider each semitone independently, some approaches attempt to model the interactions between notes, using spectral likelihood models [1,18], or music language models [3,17].…”

Polyphonic Music Sequence Transduction with Meter-Constrained LSTM Networks

“…Roughly, AMT is the task of extracting from a music recording a symbolic representation describing what notes were played and when, usually in the form of a time-pitch representation called piano-roll. AMT is a widely discussed topic, yet, unless it is constrained to a specific instrument and instrument model [1], it remains a challenging task, in particular in the case of polyphonic music: computers are far from carrying out this task as accurately as human experts [2].…”