2025 Vol.8 Jun.N03

• Title: Research on Mongolian Music Classification and Recognition Based on Deep Learning
• Name: Xiangwei Ge
• Company: Inner Mongolia Normal University, Hohhot City, Inner Mongolia Autonomous Region, 010022 China
• Abstract:

  This study proposes a deep learning based Mongolian music classification and recognition method, which combines CNN model and deep belief network (DBN) to achieve efficient feature extraction and accurate classification. Firstly, CNN is used for preprocessing and feature extraction of Mongolian music. The input data is optimized through pitch feature matrix denoising and volume calibration, and a feature sample model is generated through supervised learning training. Subsequently, the extracted features are input into an improved DBN network, which is composed of multiple stacked Restricted Boltzmann Machines (RBMs) and incorporates Dropout layers during training to prevent overfitting and enhance the model's generalization ability. DBN decomposes and reconstructs features layer by layer, transforming the original music features into higher-level abstract representations, and ultimately uses nonlinear activation functions to achieve classification and recognition of music samples. The experimental results show that this method can improve classification accuracy and the classification recognition results are more reliable.

  
  

• Keyword: Deep learning; Music classification; Music recognition; DBN network; CNN model
• DOI: 10.12250/jpciams2025090610
• Citation form: Xiangwei Ge.Research on Mongolian Music Classification and Recognition Based on Deep Learning[J]. Computer Informatization and Mechanical System,2025,Vol.8,pp.42-45

References

[1]Callum M ,Edward A F ,Pauline R .Playlists and genre: the role of music genre in Spotify’s playlists[J].Journal of Documentation,2025,81(1):1-23.

[2]Mycka J ,Mańdziuk J .Artificial intelligence in music: recent trends and challenges[J].Neural Computing and Applications,2024,37(2):1-39.

[3]ZHONG Zhipeng;WANG Hailong;SU Guibin;LIU Lin;PEI Dongmei. Music Emotion Recognition Fusion on CNN-BiLSTM and Self-Attention Model[J]. Computer Engineering and Applications, 2023,59(03):94-103.

[4]LÜXiao-jiao;ZHANG Yu-mei;YANG Hong-hong;WU Xiao-jun. A folk songs fast classification algorithm DUPSO-DSVM based on distance sorting[J]. Computer Engineering & Science, 2023,45(10):1874-1883.

[5]WEN Xinglin;BAI Tao. Design of audio emotion recognition and classification model based on improved multimodal RCNN[J]. Modern Electronics Technique, 2023,46(11):114-118.