Melodic Phrase Segmentation By Deep Neural Networks

11/14/2018
by   Yixing Guan, et al.
0

Automated melodic phrase detection and segmentation is a classical task in content-based music information retrieval and also the key towards automated music structure analysis. However, traditional methods still cannot satisfy practical requirements. In this paper, we explore and adapt various neural network architectures to see if they can be generalized to work with the symbolic representation of music and produce satisfactory melodic phrase segmentation. The main issue of applying deep-learning methods to phrase detection is the sparse labeling problem of training sets. We proposed two tailored label engineering with corresponding training techniques for different neural networks in order to make decisions at a sequential level. Experiment results show that the CNN-CRF architecture performs the best, being able to offer finer segmentation and faster to train, while CNN, Bi-LSTM-CNN and Bi-LSTM-CRF are acceptable alternatives.

READ FULL TEXT

page 6

page 7

research
08/19/2017

Arabic Multi-Dialect Segmentation: bi-LSTM-CRF vs. SVM

Arabic word segmentation is essential for a variety of NLP applications ...
research
02/06/2020

Attentional networks for music generation

Realistic music generation has always remained as a challenging problem ...
research
02/07/2019

Aspect Specific Opinion Expression Extraction using Attention based LSTM-CRF Network

Opinion phrase extraction is one of the key tasks in fine-grained sentim...
research
09/21/2022

Learning Hierarchical Metrical Structure Beyond Measures

Music contains hierarchical structures beyond beats and measures. While ...
research
11/20/2018

The Effect of Explicit Structure Encoding of Deep Neural Networks for Symbolic Music Generation

With recent breakthroughs in artificial neural networks, deep generative...
research
08/21/2018

Language Identification in Code-Mixed Data using Multichannel Neural Networks and Context Capture

An accurate language identification tool is an absolute necessity for bu...

Please sign up or login with your details

Forgot password? Click here to reset