Music genre classification using convolutional neural network
| dc.contributor.advisor | Singh, B.K. | |
| dc.contributor.author | Bhandari, Neema | |
| dc.date.accessioned | 2021-08-18T10:41:55Z | |
| dc.date.available | 2021-08-18T10:41:55Z | |
| dc.date.issued | 2021-02 | |
| dc.description.abstract | Music applications are one of the most used applications in the world. Music Genre Classification (MGC) has been gaining attention with the rise of digital music and it is a useful tool for semantic information to music tracks in offline and online music collections. A music genre refers to a specific class of music with a set of common properties. A mere perception of the music of that class can help one to distinguish it from other classes. Musical genre classification is a promising yet difficult task in the field of musical information retrieval. To determine the genre of a song it has to be distinguished by its unique audio features so that its contents can be analyzed with respect to the produced wave signals. A single Music Genre is a set of different features that combined with a specific pattern of rhythm, melody, harmony, instruments, mood and attitude, lyrics and language. In the Western music there has been much work done in the area of automatic tagging genre recognition, classification and comparative studies as compare to the Indian music. As a widely used feature in genre classification systems, Mel-frequency cepstral coefficients (MFCC) is typically believed to encode timbral information, since it represents shortduration musical textures. In this thesis, we investigate the invariance of MFCC and show that MFCCs in fact encode both timbral and key information. Convolutional Neural Networks have additional layers for edge detection that make them well suited for classification problems. For the convolutional base classification, we used a common pattern one is a stack of Conv2D and second one is MaxPooling2D layers. In this research work we use six semi classical Indian music genre like Abhang, Bhajan, Kajari, Qawwali, Tappa and Thumri. | en_US |
| dc.identifier.uri | https://krishikosh.egranth.ac.in/handle/1/5810172340 | |
| dc.keywords | music genre classification, convolution, neural networks, digital technology | en_US |
| dc.language.iso | English | en_US |
| dc.pages | 67 | en_US |
| dc.publisher | G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand) | en_US |
| dc.research.problem | Music Genre Classification | en_US |
| dc.sub | Computer Engineering | en_US |
| dc.theme | Neural Networks | en_US |
| dc.these.type | M.Tech. | en_US |
| dc.title | Music genre classification using convolutional neural network | en_US |
| dc.type | Thesis | en_US |