Automatic Music Highlight Extraction using Convolutional Recurrent Attention Networks

2017/12/16 by Jung-Woo Ha, Adrian Kim, Ha, Jung-Woo +7 · 1 citation
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Multimedia (cs.MM) #Music and Audio Processing #Sound (cs.SD) #Speech and Audio Processing #Video Analysis and Summarization

paper · pdf · doi:10.48550/arxiv.1712.05901

openalex publication_date 2017/12/16 · openalex created_date 2018/01/05 · openalex updated_date 2026/07/28

Abstract

Music highlights are valuable contents for music services. Most methods focused on low-level signal features. We propose a method for extracting highlights using high-level features from convolutional recurrent attention networks (CRAN). CRAN utilizes convolution and recurrent layers for sequential learning with an attention mechanism. The attention allows CRAN to capture significant snippets for distinguishing between genres, thus being used as a high-level feature. CRAN was evaluated on over 32,000 popular tracks in Korea for two months. Experimental results show our method outperforms three baseline methods through quantitative and qualitative evaluations. Also, we analyze the effects of attention and sequence information on performance.

Citations

• Adam: A Method for Stochastic Optimization
• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• Convolutional Recurrent Neural Networks for Music Classification
• Dual Attention Networks for Multimodal Reasoning and Matching

Cited by

• Constructing a Highlight Classifier with an Attention Based LSTM Neural Network

Related

• Convolutional Recurrent Neural Networks for Music Classification
• Multi-scale temporal-frequency attention for music source separation
• MuSLCAT: Multi-Scale Multi-Level Convolutional Attention Transformer for Discriminative Music Modeling on Raw Waveforms
• Visualizing and Understanding Self-attention based Music Tagging
• Attention-guided Spectrogram Sequence Modeling with CNNs for Music Genre Classification