A real-time peak discovering method for audio fingerprinting

  • Authors:

  • Tao Jiang;Rihui Wu;Jiahong Li;Kang Xiang;Feng Dai

  • Affiliations:

  • Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS);University of Chinese Academy of Sciences, Beijing, China;University of Chinese Academy of Sciences, Beijing, China;University of Chinese Academy of Sciences, Beijing, China;Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS)

  • Venue:
  • Proceedings of the Fifth International Conference on Internet Multimedia Computing and Service
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Discovering peaks is a premise of audio fingerprinting algorithms for audio information retrieval, which focus on spectrogram peak pairs. In this paper, we discuss finding peaks in a two-direction scanning method which use a dynamic threshold vector to find local maximal point as peaks. And then an improved method using a slide window with two grids is proposed for discovering peaks at real-time. In this method, two key steps are executed alternately: scanning and sliding. In each scanning step, new peaks are found. When the slide window method is used, experiments on a database with 400,000 songs show that the average query time is shorten and the recall rate is slightly decreased.