Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 43387 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Sameti, Hossein
- Abstract:
- Measuring emotions of music is one of the methods to determine music content. Music emotion detection is applicable in music retrieval, recognition of music genre and also music data management softwares. Music emotion is considered in different sciences such as physiology, psychology, musicology and engineering. First, we collected a database of different types of music with various emotions. These data have been labeled according to their emotions. In this project, four emotions (Angry, happy, relax and sad) have been used as labels based on Thayer’s two dimension emotion model. There are two basic steps for music emotion recognition similar to other recognition systems: Feature extraction and classification. In the first step, significant and useful features in music were extracted. These features are classified into energy-related, rhythm and timbre. Then, three feature selection methods were applied to our feature set to find the best features. In the second step, two training models have been used to classify music into four emotions. The first model is a two level algorithm based on support vector machine that increased accuracy rate about 12% in comparison with basic algorithms without feature selection. Second model is a semi-supervised algorithm based on sparse representation that improved accuracy rate about 14% in comparison with applying only sparse representation. Maximum accuracy of proposed algorithms was 87.85%.
- Keywords:
- Classification ; Feature Extraction ; Information Retrieval ; Music Emotion
Digital Object List-
محتواي پايان نامه - view
Bookmark