From Note-Level to Chord-Level Neural Network Models for Voice Separation in Symbolic Music
This work addresses the problem of automatically separating voices in music for applications in music analysis and generation, representing an incremental improvement over existing methods.
The paper tackles voice separation in symbolic music by defining a flexible task that allows voices to diverge and converge, and trains neural networks at note-level and chord-level to assign notes to voices. Both models surpass a strong baseline in extracting consecutive within-voice note pairs, with the chord-level model performing slightly better, and they also outperform previous approaches on Bach music.
Music is often experienced as a progression of concurrent streams of notes, or voices. The degree to which this happens depends on the position along a voice-leading continuum, ranging from monophonic, to homophonic, to polyphonic, which complicates the design of automatic voice separation models. We address this continuum by defining voice separation as the task of decomposing music into streams that exhibit both a high degree of external perceptual separation from the other streams and a high degree of internal perceptual consistency. The proposed voice separation task allows for a voice to diverge to multiple voices and also for multiple voices to converge to the same voice. Equipped with this flexible task definition, we manually annotated a corpus of popular music and used it to train neural networks that assign notes to voices either separately for each note in a chord (note-level), or jointly to all notes in a chord (chord-level). The trained neural models greedily assign notes to voices in a left to right traversal of the input chord sequence, using a diverse set of perceptually informed input features. When evaluated on the extraction of consecutive within voice note pairs, both models surpass a strong baseline based on an iterative application of an envelope extraction function, with the chord-level model consistently edging out the note-level model. The two models are also shown to outperform previous approaches on separating the voices in Bach music.