A Combination of Hand-Crafted and Hierarchical High-Level Learnt Feature Extraction for Music Genre Classification
نویسندگان
چکیده
In this paper, we propose a new approach for automatic music genre classification which relies on learning a feature hierarchy with a deep learning architecture over hand-crafted feature extracted from an audio signal. Unlike the state-of-the-art approaches, our scheme uses an unsupervised learning algorithm based on Deep Belief Networks (DBN) learnt on block-wise MFCC (that we treat as 2D images), followed by a supervised learning algorithm for fine-tuning the extracted features. Experiments performed on the GTZAN dataset show that the proposed scheme clearly outperforms the state-of-the-art approaches.
منابع مشابه
شناسایی خودکار سبک موسیقی
Nowadays, automatic analysis of music signals has gained a considerable importance due to the growing amount of music data found on the Web. Music genre classification is one of the interesting research areas in music information retrieval systems. In this paper several techniques were implemented and evaluated for music genre classification including feature extraction, feature selection and m...
متن کاملLocal-feature-map Integration Using Convolutional Neural Networks for Music Genre Classification
A map-based approach, which treats 2-dimensional acoustic features using image analysis, has recently attracted attention in music genre classification. While this is successful at extracting local music-patterns compared with other frame-based methods, in most works the extracted features are not sufficient for music genre classification. In this paper, we focus on appropriate feature extracti...
متن کاملComparing Features for the Automatic Classification of Vocal Music
Traditional approaches to the task of automatic classification of music by genre generally focus on the use of note-duration symbols to represent musical content. As well, most studies generally extract information from a combination of instruments. This paper compares the traditional method of using note-duration symbols with another method based on time slice representations of the music to s...
متن کاملHigh-Level Music Descriptor Extraction Algorithm Based on Combination of Multi-Channel CNNs and LSTM
Although Convolutional Neural Networks (CNNs) and Long Short Term Memory (LSTM) have yielded impressive performances in a variety of Music Information Retrieval (MIR) tasks, the complementarity among the CNNs of different architectures and that between CNNs and LSTM are seldom considered. In this paper, multichannel CNNs with different architectures and LSTM are combined into one unified archit...
متن کاملUnsupervised learning of low-level audio features for music similarity estimation
While there is an enormous amount of music data available, the field of music analysis almost exclusively uses manually designed features. In this work we learn features from music data in a completely unsupervised way and evaluate them on a musical genre classification task. We achieve results very close to state-of-the-art performance which relies on highly hand-tuned feature extractors.
متن کامل