Dissertation > Excellent graduate degree dissertation topics show

Research on the Key Technology of the Content-based Music Retrieval

Author: SuQiYuan
Tutor: HanGuoQiang；YuanHuaQiang
School: South China University of Technology
Course: Applied Computer Technology
Keywords: Query by Humming Music melody Pitch Detection Approximate match DTW algorithm
CLC: TP391.3
Type: Master's thesis
Year: 2011
Downloads: 65
Quote: 0
Read: Download Dissertation

Abstract


With the rapid development of computer network technology and multimedia technology, the amount of music data information has skyrocketed. How quickly and accurately from the vast music data to find the music you want has become a hot research topic in modern information retrieval field. Traditional text-based music retrieval only by entering the song name or artist name and other information retrieval, has been unable to meet the demand for music retrieval. Content-based retrieval allows the user to retrieve the desired songs by humming the form, even if a user forget the song name or artist name and other information, will be able to find the song you want as long as he can sing a part of the song snippet. This new type of human-computer interface for music retrieval has great practical significance, academics, developers, and the vast majority of users are more and more attention. In this paper, the National Natural Science Foundation project three-dimensional drama animation synthesis technology (NO.U0935003) \The matching of the melody. The essential characteristics of the analysis of music and music melody based on the method proposed the melodic features an improved integrated use of sound elevation and cadence than binary notation. This method can more accurately represent a melody in the music signal characteristics, and having a better retrieval accuracy. The music melody feature extraction, through traditional gene cycle detection technology, found that the traditional gene cycle detection technology there are a variety of defects. Multi-level wavelet transform the genetic cycle detection for these defects article, the first of the original speech signal, the higher several levels approaching signal weighted sum of processing, and then combine the autocorrelation function and average magnitude difference function synthesized signal for pitch detection. And male and female voices signals using different weighting coefficients, so that the two signals can be obtained to the smooth melodic contour curve. In this paper, a numerical index melody matching retrieval module, melody the pitch difference between the information in the feature library indexing and matching algorithms designed on the basis of a melody. Finally, we designed a simple humming-based music retrieval system, detailed experiments proved the effectiveness and feasibility of the proposed method.

Related Dissertations

  1. A Music Retrieval System Base on Humming,TN912.3
  2. MFCC -based speech recognition system to improve research and design,TN912.34
  3. Research on the Algorithm of Speeth Pitch Detection,TN912.3
  4. Study of Extraction and Optimization Characteristic Parameters in Speaker Recognition,TN912.34
  5. Exploring the Effects of Melody, Harmony and Rhythm on Valence or Pleasantness of Chopin’s Musical Works,J624.1
  6. Research on Approximate Tiwg Pattern Matching of XML Based on DTD,TP311.13
  7. The Design of Computer Software in Piano Tuning,TP368.1
  8. Research of the Speech Recognition Technology Based on DSP,TN912.34
  9. Research on Speaker Characteristic Parameter Extraction in the Noise Environment,TN912.3
  10. The Clustering of Time-Series Data Based on LB_Hust Distance Caculation,TP311.13
  11. Baby crying sound feature analysis and recognition,TN912.34
  12. Research and Application of Feature Clustering Method Based on Music Melody,TP311.13
  13. Research and Implementation on Content-Based Music Retrieval Technology,TP391.3
  14. Study on Content-based Music Recognition,TP391.3
  15. Dynamic gesture recognition research based on the optical flow computation and DTW algorithm,TP391.41
  16. Voice Application System Technology Research Based on Voice Sample Matching,TP391.42
  17. The Study of the 20th Century Music Melody Hear Training,J613.1
  18. Study of Pitch Detection Algorithm and the Application in Dialect Identification,TP391.42
  19. Humming - based music retrieval technology research,TP391.3
  20. Research on Music Retrieval Based on Content,TP391.3

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Retrieval machine
© 2012 www.DissertationTopic.Net  Mobile