• Corpus ID: 12211154

Deep Learning Techniques for Music Generation - A Survey

@article{Briot2017DeepLT,
  title={Deep Learning Techniques for Music Generation - A Survey},
  author={Jean-Pierre Briot and Ga{\"e}tan Hadjeres and François Pachet},
  journal={ArXiv},
  year={2017},
  volume={abs/1709.01620}
}
This paper is a survey and an analysis of different ways of using deep learning (deep artificial neural networks) to generate musical content. [] Key Method These systems are described and are used to exemplify the various choices of objective, representation, architecture, challenge and strategy. The last section includes some discussion and some prospects.

Figures and Tables from this paper

LSTM Based Music Generation
TLDR
A goal is to generate musical notes that will follow a given input primer sequence such that the entire generated musical sequence sounds continuous and melodious.
Interactive Control of Explicit Musical Features in Generative LSTM-based Systems
TLDR
A very basic LSTM-based architecture is developed that can compose music that corresponds to user-provided values of rhythm density and pitch height/register and is suitable for generating music the features of which are defined in real-time and/or interactively.
SynthNet: Learning to Synthesize Music End-to-End
TLDR
It is concluded that mappings between musical notes and the instrument timbre can be learned directly from the raw audio coupled with the musical score, in binary piano roll format.
Music Generation with Deep Neural Networks Using Flattened Multi-Channel Skip-3 Softmax and Cross-Entropy
TLDR
An investigation of convolutional neural networks as a means of generating human-plausible, goal-oriented music, specifically pop melodies shows some evidence of rhythmic and harmonic patterns, but lack melodic elements.
Deep Neural Networks for Generating Music
TLDR
A Tied Parallel Network is used, which is a combination of a recurrent and a feedforward network, and a model to generate songs in such a manner that the rhythm is controlled accordingly to the speed of a person is modified.
Deep learning for music generation: challenges and directions
TLDR
Some limitations of a direct application of deep learning to music generation are selected and why the issues are not fulfilled and how to address them by possible approaches are analyzed.
Deep Learning for Expressive Music Generation
TLDR
Results suggest that expressive elements such as duration and velocity are key aspects in a music composition expression, making the ones who include these preferable to non-expressive ones.
A Comprehensive Survey on Deep Music Generation: Multi-level Representations, Algorithms, Evaluations, and Future Directions
TLDR
This paper attempts to provide an overview of various composition tasks under different music generation levels, covering most of the currently popular music generation tasks using deep learning.
RaveForce : A Deep Reinforcement Learning Environment for Music Generation
TLDR
The proposed method, RaveForce, can be used to search new synthesis parameters for a specific timbre of an electronic music note or loop with deep reinforcement learning algorithms.
Next Bar Predictor: An Architecture in Automated Music Generation
TLDR
The Next Bar Predictor is a generative model that creates melody one bar at a time using the previous bar as basis to generate aesthetically pleasing melodies, and based on the evaluation by human listeners, the melodies generated by these models are more realistic and pleasing than those of the Midinet.
...
...

References

SHOWING 1-10 OF 241 REFERENCES
A Unit Selection Methodology for Music Generation Using Deep Neural Networks
TLDR
This work describes a generative model that combines a deep structured semantic model (DSSM) with an LSTM to predict the next unit, where units consist of four, two, and one measures of music.
Music Generation by Deep Learning - Challenges and Directions
TLDR
Some limitations of a direct application of deep learning to music generation are selected, why the issues are not fulfilled and how to address them by possible approaches are analyzed.
DeepJ: Style-Specific Music Generation
TLDR
This paper introduces DeepJ - an end-to-end generative model that is capable of composing music conditioned on a specific mixture of composer styles, and innovations include methods to learn musical style and music dynamics.
Conditioning Deep Generative Raw Audio Models for Structured Automatic Music
TLDR
This paper considers a Long Short Term Memory network to learn the melodic structure of different styles of music, and then uses the unique symbolic generations from this model as a conditioning input to a WaveNet-based raw audio generator, creating a model for automatic, novel music.
MidiNet: A Convolutional Generative Adversarial Network for Symbolic-Domain Music Generation
TLDR
This work proposes a novel conditional mechanism to exploit available prior knowledge, so that the model can generate melodies either from scratch, by following a chord sequence, or by conditioning on the melody of previous bars, making it a generative adversarial network (GAN).
Feature learning and deep architectures: new directions for music informatics
TLDR
This work critically review the standard approach to music signal analysis and identifies three specific deficiencies to current methods: hand-crafted feature design is sub-optimal and unsustainable, the power of shallow architectures is fundamentally limited, and short-time analysis cannot encode musically meaningful structure.
Musical Audio Synthesis Using Autoencoding Neural Nets
TLDR
An interactive musi- cal audio synthesis system that uses feedforward artificial neural networks for musical audio synthesis, rather than discriminative or regression tasks, and allows one to interact directly with the parameters of the model and generate musical audio in real time.
Modelling Symbolic Music: Beyond the Piano Roll
TLDR
A representation which reduces polyphonic music to a univariate categorical sequence is introduced, which is able to apply state of the art natural language processing techniques, namely the long short-term memory sequence model.
Sampling Variations of Sequences for Structured Music Generation
TLDR
This work presents an approach to generate structured sequences, based on a mechanism for sampling efficiently variations of musical sequences, that can be used to implement composition strategies that enforce arbitrary structure on a musical lead sheet generation problem.
Music transcription modelling and composition using deep learning
TLDR
This work builds and train LSTM networks using approximately 23,000 music transcriptions expressed with a high-level vocabulary (ABC notation), and uses them to generate new transcriptions to create music transcription models useful in particular contexts of music composition.
...
...