-
Notochord: a Flexible Probabilistic Model for Real-Time MIDI Performance
Authors:
Victor Shepardson,
Jack Armitage,
Thor Magnusson
Abstract:
Deep learning-based probabilistic models of musical data are producing increasingly realistic results and promise to enter creative workflows of many kinds. Yet they have been little-studied in a performance setting, where the results of user actions typically ought to feel instantaneous. To enable such study, we designed Notochord, a deep probabilistic model for sequences of structured events, an…
▽ More
Deep learning-based probabilistic models of musical data are producing increasingly realistic results and promise to enter creative workflows of many kinds. Yet they have been little-studied in a performance setting, where the results of user actions typically ought to feel instantaneous. To enable such study, we designed Notochord, a deep probabilistic model for sequences of structured events, and trained an instance of it on the Lakh MIDI dataset. Our probabilistic formulation allows interpretable interventions at a sub-event level, which enables one model to act as a backbone for diverse interactive musical functions including steerable generation, harmonization, machine improvisation, and likelihood-based interfaces. Notochord can generate polyphonic and multi-track MIDI, and respond to inputs with latency below ten milliseconds. Training code, model checkpoints and interactive examples are provided as open source software.
△ Less
Submitted 18 March, 2024;
originally announced March 2024.
-
A variational autoencoder for music generation controlled by tonal tension
Authors:
Rui Guo,
Ivor Simpson,
Thor Magnusson,
Chris Kiefer,
Dorien Herremans
Abstract:
Many of the music generation systems based on neural networks are fully autonomous and do not offer control over the generation process. In this research, we present a controllable music generation system in terms of tonal tension. We incorporate two tonal tension measures based on the Spiral Array Tension theory into a variational autoencoder model. This allows us to control the direction of the…
▽ More
Many of the music generation systems based on neural networks are fully autonomous and do not offer control over the generation process. In this research, we present a controllable music generation system in terms of tonal tension. We incorporate two tonal tension measures based on the Spiral Array Tension theory into a variational autoencoder model. This allows us to control the direction of the tonal tension throughout the generated piece, as well as the overall level of tonal tension. Given a seed musical fragment, stemming from either the user input or from directly sampling from the latent space, the model can generate variations of this original seed fragment with altered tonal tension. This altered music still resembles the seed music rhythmically, but the pitch of the notes are changed to match the desired tonal tension as conditioned by the user.
△ Less
Submitted 14 October, 2020; v1 submitted 13 October, 2020;
originally announced October 2020.
-
Midi Miner -- A Python library for tonal tension and track classification
Authors:
Rui Guo,
Dorien Herremans,
Thor Magnusson
Abstract:
We present a Python library, called Midi Miner, that can calculate tonal tension and classify different tracks. MIDI (Music Instrument Digital Interface) is a hardware and software standard for communicating musical events between digital music devices. It is often used for tasks such as music representation, communication between devices, and even music generation [5]. Tension is an essential ele…
▽ More
We present a Python library, called Midi Miner, that can calculate tonal tension and classify different tracks. MIDI (Music Instrument Digital Interface) is a hardware and software standard for communicating musical events between digital music devices. It is often used for tasks such as music representation, communication between devices, and even music generation [5]. Tension is an essential element of the music listening experience, which can come from a number of musical features including timbre, loudness and harmony [3]. Midi Miner provides a Python implementation for the tonal tension model based on the spiral array [1] as presented by Herremans and Chew [4]. Midi Miner also performs key estimation and includes a track classifier that can disentangle melody, bass, and harmony tracks. Even though tracks are often separated in MIDI files, the musical function of each track is not always clear. The track classifier keeps the identified tracks and discards messy tracks, which can enable further analysis and training tasks.
△ Less
Submitted 26 May, 2020; v1 submitted 3 October, 2019;
originally announced October 2019.
