Abstract
Because of the difficulty of music feature recognition due to the complex and varied music theory knowledge influenced by music specialization, we designed a music feature recognition system based on Internet of Things (IoT) technology. The physical sensing layer of the system places sound sensors at different locations to collect the original music signals and uses a digital signal processor to carry out music signal analysis and processing. The network transmission layer transmits the completed music signals to the music signal database in the application layer of the system. The music feature analysis module of the application layer uses a dynamic time regularization algorithm to obtain the maximum similarity between the test template and the reference. The music feature analysis module of the application layer uses the dynamic time regularization algorithm to obtain the maximum similarity between the test template and the reference template to realize the feature recognition of the music signal and determine the music pattern and music emotion corresponding to the music feature content according to the recognition result. The experimental results show that the system operates stably, can capture high-quality music signals, and can correctly identify music style features and emotion features. The results of this study can meet the needs of composers’ assisted creation and music researchers’ analysis of a large amount of music data, and the results can be further transferred to deep music learning research, human-computer interaction music creation, application-based music creation, and other fields for expansion.
Highlights
E current music recognition system has not been applied on a large scale due to the lack of an overall system framework design that facilitates the enhancement of performance information. e advent of Internet of ings (IoT) technology has made the implementation of music feature recognition systems possible [2]. e IoT technology realizes the intelligent collection, processing, and division of music signals through real-time information transmission between wired/wireless networks, which has the advantages of comprehensive perception, reliable transmission, and convenience [3]. e design of music feature recognition system is based on IoT technology to realize the perception, transmission, and recognition of music signals
The structural characteristics of music styles can vary with the styles, and even the same person can sing differently with different ranges when performing different kinds of music styles [6]. All these factors make it very difficult to extract the features of music signals, which in turn makes it difficult to improve the accuracy of music style recognition. erefore, recognizing different music styles has recently received a lot of attention and has been developed rapidly
A new deep learning framework is proposed for music recognition based on convolutional recurrent hashing, which adds RNN layer and hash layer after CNN, the encoding obtained from the hash layer is input to the classifier for classification, and SoftMax classification loss is used as one of the optimization objectives. e hash function that preserves similarity is first learned by using a convolutional feature map containing spatial details and semantic information, which is extracted from multiple convolutional layers by inputting the preprocessed music signal spectrogram into a pretrained CNN and subsequently generating effective hash codes using a multilayer RNN, which can directly use the convolutional feature map as input to maintain its spatial structure
Summary
Music feature recognition is based on the art of speech recognition development, where the audio signal is used to obtain the musical content and the musical features such as musical patterns and emotions [1]. e study of music feature recognition involves many aspects such as psychoacoustics, instrument analysis, and music theory knowledge. Many developers only pursue the beauty of music players, with a very large number of functions and comprehensive processing capabilities, while the average user cannot use these functions, while increasing the difficulty of operation; the limited storage space of cell phones and music players take up too much memory of cell phones, increasing the burden of cell phones, resulting in system slowdown; Complexity sometimes there will be a crash situation, causing unnecessary trouble. The structural characteristics of music styles can vary with the styles, and even the same person can sing differently with different ranges when performing different kinds of music styles [6] All these factors make it very difficult to extract the features of music signals, which in turn makes it difficult to improve the accuracy of music style recognition. E effect of quantization error is considered in the whole deep hash model framework and the hash layer output. e quantization error is generated when the binary hash code is added to the optimization objective to obtain a more expressive hash code
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
