Proceedings of the 6th Conference on Sound and Music Technology (CSMT)

1. Front Matter

   Pages i-viii

   PDF

2. Music Processing and Music Information Retrieval

   1. Front Matter

      Pages 1-1

      PDF

   2. A Novel Singer Identification Method Using GMM-UBM

      • Xulong Zhang, Yiliang Jiang, Jin Deng, Juanjuan Li, Mi Tian, Wei Li

      Pages 3-14

   3. A Practical Singing Voice Detection System Based on GRU-RNN

      • Zhigao Chen, Xulong Zhang, Jin Deng, Juanjuan Li, Yiliang Jiang, Wei Li

      Pages 15-25

   4. Multimodel Music Emotion Recognition Using Unsupervised Deep Neural Networks

      • Jianchao Zhou, Xiaoou Chen, Deshun Yang

      Pages 27-39

   5. Music Summary Detection with State Space Embedding and Recurrence Plot

      • Yongwei Gao, Yichun Shen, Xulong Zhang, Shuai Yu, Wei Li

      Pages 41-51

   6. Constructing a Multimedia Chinese Musical Instrument Database

      • Xiaojing Liang, Zijin Li, Jingyu Liu, Wei Li, Jiaxing Zhu, Baoqiang Han

      Pages 53-60

3. Acoustic Sound Processing and Analysis

   1. Front Matter

      Pages 61-61

      PDF

   2. Bird Sound Detection Based on Binarized Convolutional Neural Networks

      • Jianan Song, Shengchen Li

      Pages 63-71

   3. Adaptive Consistent Dictionary Learning for Audio Declipping

      • Penglong Wu, Xia Zou, Meng Sun, Li Li, Xingyu Zhang

      Pages 73-83

   4. A Comparison of Attention Mechanisms of Convolutional Neural Network in Weakly Labeled Audio Tagging

      • Yuanbo Hou, Qiuqiang Kong, Shengchen Li

      Pages 85-96

4. Music Steganography

   1. Front Matter

      Pages 97-97

      PDF

   2. A Standard MIDI File Steganography Based on Music Perception in Note Duration

      • Lei Guan, Yinji Jing, Shengchen Li, Ru Zhang

      Pages 99-107

This book discusses the use of advanced techniques to produce and understand music in a digital way. It gathers the first-ever English-language proceedings of the Conference on Sound and Music Technology (CSMT), which was held in Xiamen, China in 2018. As a leading event, the CSMT reflects the latest advances in acoustic and music technologies in China.  
Sound and technology are more closely linked than most people assume. For example, signal-processing methods form the basis of music feature extraction, while mathematics provides an objective means of representing current musicological theories and discovering new ones. Moreover, machine-learning methods include popular deep learning algorithms and are used in a broad range of contexts, from discovering patterns in music features to producing music. As these proceedings demonstrate, modern technologies not only offer new ways to create music, but can also help people perceive sound in innovative new ways.

• audio signal processing
• computer audition
• composition
• music production
• sound design
• Audio Information Security
• Auditory Psychology

• School of Computer Science and Technology, Fudan University, Shanghai, China

  Wei Li

• Beijing University of Posts and Telecommunications, Beijing, China

  Shengchen Li

• Institute of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, China

  Xi Shao

• China Conservatory of Music, Beijing, China

  Zijin Li

Policies and ethics