PopMNet: Generating structured pop music melodies using neural networks

Wu, Jian; Liu, Xiaoguang; Hu, Xiaolin; Zhu, Jun

doi:10.1016/j.artint.2020.103303

Cited by 36 publications

(18 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Further discussion can ask the meaning of an aesthetic work of art. Computers may have the capacity to produce images or music, for example, that seems or sound appealing [31,32].…”

Section: Discussionmentioning

confidence: 99%

From Science Fiction to Science Facts

Satwiko¹,

Michelle²

2022

Advances in Social Science, Education and Humanities Research

View full text Add to dashboard Cite

The current development of artificial intelligence (AI) may encourage the digital architect emergence. It is an artificial-intelligence-powered software that approaches the architectural design process from upstream to downstream in an integrated and comprehensive manner and works autonomously. This paper reviews research in the field of AI to discuss the development of the concept of the digital architect. The result explained the presence of digital architect software as a necessity to bring more efficient and holistic architectural design. Furthermore, we concluded that the final development of the digital architect might be only a matter of time, considering the advancements of intelligent robot technology.

show abstract

“…Further discussion can ask the meaning of an aesthetic work of art. Computers may have the capacity to produce images or music, for example, that seems or sound appealing [31,32].…”

Section: Discussionmentioning

confidence: 99%

From Science Fiction to Science Facts

Satwiko¹,

Michelle²

2022

Advances in Social Science, Education and Humanities Research

View full text Add to dashboard Cite

show abstract

“…These models enable free generation and motif continuation, but it is difficult to control the generated content. StructureNet [3], PopMNet [24] and Racchmaninof [19] are more closely related to our work in that they introduce explicit models for music structure.…”

Section: Related Workmentioning

confidence: 99%

Controllable deep melody generation via hierarchical music structure representation

Dai¹,

Jin²,

gomes³

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent advances in deep learning have expanded possibilities to generate music, but generating a customizable full piece of music with consistent long-term structure remains a challenge. This paper introduces MusicFrameworks, a hierarchical music structure representation and a multi-step generative process to create a full-length melody guided by long-term repetitive structure, chord, melodic contour, and rhythm constraints. We first organize the full melody with section and phrase-level structure. To generate melody in each phrase, we generate rhythm and basic melody using two separate transformer-based networks, and then generate the melody conditioned on the basic melody, rhythm and chords in an auto-regressive manner. By factoring music generation into sub-problems, our approach allows simpler models and requires less data. To customize or add variety, one can alter chords, basic melody, and rhythm structure in the music frameworks, letting our networks generate the melody accordingly. Additionally, we introduce new features to encode musical positional information, rhythm patterns, and melodic contours based on musical domain knowledge. A listening test reveals that melodies generated by our method are rated as good as or better than human-composed music in the POP909 dataset about half the time.

show abstract

“…Other areas where RNNs have been successful include deep learning [98], smart network management [14,31,36,37,40], emergency management and cyberphysical systems [42,43], the dynamic management of Cloud and Fog services [21,22,90], the use of machine learning in smart search [75,76] and network routing including the use of Software Defined Networks [5,19,20,29,30,38,91]. The CNN has also been successfully used in many fields [1,58], including for Magnetic Resonance Image reconstruction [94], automatic road segmentation [57], music generation [93] and relation extraction from plain text [45]. This wide usage and success of both the RNN and the CNN in a variety of applications justifies their use for Software Vulnerability Prediction in this paper, where text data processing and dimensionality reduction is carried out with a small CNN, and the RNN is used as a model that bonds both parts of the analysis.…”

Section: Scope Of the Present Workmentioning

confidence: 99%

Modelling, Analysis, and Simulation of Computer and Telecommunication Systems

Filus

Siavvas

Domańska

et al. 2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Software vulnerability prediction is an important and active area of research where new methods are needed to build accurate and efficient tools that can identify security issues. Thus we propose an approach based on mixed features that combines text mining features and the features generated using a Static Code Analyzer. We use a Random Neural Network as a bonding model that combines the text analysis that is carried out on software using a Convolutional Neural Network, and the outputs of Static Code Analysis. The proposed approach was evaluated on commonly used datasets and led to 97% training accuracy, and 93%-94% testing accuracy, with a 1% reduction in false positives with respect to previously published results on similar data sets.

show abstract

PopMNet: Generating structured pop music melodies using neural networks

Cited by 36 publications

References 6 publications

From Science Fiction to Science Facts

From Science Fiction to Science Facts

Controllable deep melody generation via hierarchical music structure representation

Modelling, Analysis, and Simulation of Computer and Telecommunication Systems

Contact Info

Product

Resources

About