DocumentCode
1633838
Title
Pitch recognition based on intelligent neural network system
Author
Li, Minyu ; Li, Tao
Author_Institution
Dept. of Comput., Sichuan Univ., Chengdu, China
Volume
2
fYear
2004
Firstpage
1081
Abstract
Pitch detection may be used in content-based classification and retrieval of music and audio. Most pitch detection algorithms are based on the relationship between frequency and note; however, when the pitch is hummed by an untrained person, the accuracy of pitch recognition can obviously decline. A new intelligent pitch recognition (IPR) method based on intelligent neural networks is presented. Using the ideas of intelligent neural networks, a complex task of pitch recognition can be divided into several simple ones, each of which can be easily implemented by some simple intelligent neuron. Then a large network built out of those intelligent neurons can solve the original, complex problem and the work is much easier than with traditional neural network methods. Experimental results show it is a good way for pitch recognition.
Keywords
audio signal processing; content-based retrieval; feature extraction; learning (artificial intelligence); neural nets; pattern classification; signal classification; audio; content-based classification; content-based retrieval; feature extraction; intelligent neural network system; intelligent pitch recognition; music; pitch detection; Detection algorithms; Discrete wavelet transforms; Frequency; Intellectual property; Intelligent networks; Intelligent systems; Music information retrieval; Neural networks; Neurons; Time domain analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
Communications, Circuits and Systems, 2004. ICCCAS 2004. 2004 International Conference on
Print_ISBN
0-7803-8647-7
Type
conf
DOI
10.1109/ICCCAS.2004.1346364
Filename
1346364
Link To Document