Using Autonomous Agents to Improvise Music Compositions in Real-Time

This paper outlines an approach to real-time music generation using melody and harmony focused agents in a process inspired by jazz improvisation. A harmony agent employs a Long Short-Term Memory (LSTM) artificial neural network trained on the chord progressions of 2986 jazz ‘standard’ compositions using a network structure novel to chord sequence analysis. The melody agent uses a rule-based system of manipulating provided, pre-composed melodies to improvise new themes and variations. The agents take turns in leading the direction of the composition based on a rating system that rewards harmonic consistency and melodic flow. In developing the multi-agent system it was found that implementing embedded spaces in the LSTM encoding process resulted in significant improvements to chord sequence learning.

[1]  Todd J. Hostager,et al.  Jazz as a Process of Organizational Innovation , 1988 .

[2]  François Pachet,et al.  Assisted Lead Sheet Composition Using FlowComposer , 2016, CP.

[3]  Bhuvana Ramabhadran,et al.  Using continuous lexical embeddings to improve symbolic-prosody prediction in a text-to-speech front-end , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[4]  J. Schmidhuber,et al.  A First Look at Music Composition using LSTM Recurrent Neural Networks , 2002 .

[5]  John A. Biles,et al.  GenJam: evolution of a jazz improviser , 2001 .

[6]  Berner Lindström,et al.  Compositional Strategies in Computer-Based Music-Making , 1996, British Journal of Music Education.

[7]  Davide Morelli,et al.  Experience-Driven Procedural Music Generation for Games , 2012, IEEE Transactions on Computational Intelligence and AI in Games.

[8]  Anil Gaba,et al.  Coda : Creativity and Improvisation in Jazzand Organizations: Implications Fororganizational Learning , 1998 .

[9]  Mark Levine,et al.  The Jazz Theory Book , 1995 .

[10]  Jürgen Schmidhuber,et al.  Learning Precise Timing with LSTM Recurrent Networks , 2003, J. Mach. Learn. Res..

[11]  Arne Eigenfeldt,et al.  Realtime Generation of Harmonic Progressions Using Constrained Markov Selection , 2010, ICCC.

[12]  John A. Biles,et al.  GenJam in Transition: from Genetic Jammer to Generative Jammer , 2002 .

[13]  François Pachet,et al.  Imitative Leadsheet Generation with User Constraints , 2014, ECAI.

[14]  Philip N. Johnson-Laird,et al.  How Jazz Musicians Improvise , 2002 .