A New Feature for Musical Genre Classification of MPEG-4 TwinVQ Audio Data
スポンサーリンク
概要
- 論文の詳細を見る
This paper proposes a new musical feature to classify MPEG-4 TwinVQ compressed data into musical genre without decoding to audio signals. To extract the musical feature, we use the LSP (Line Spectrum Pair) parameters directly extracted from a bitstream without any computation, and the Discrete Wavelet Transform (DWT). We experimented on 2,196 compressed music data collected from 10 musical genres and evaluated the performance of the musical feature for musical genre classification. The maximum of average correct ratio for musical genre classification was 81.7%. Experiment showed that the musical feature had very good performance for musical genre classification in the compressed domain of MPEG-4 TwinVQ audio compression.
著者
-
Hoshi Mamoru
Graduate School Of Information Systems The University Of Electro-communications
-
Ohmori Tadashi
Graduate School Of Information Systems The University Of Electro-communications
-
Kobayakawa Michihiro
Graduate School Of Information Systems The University Of Electro-communications
-
Kobayakawa Michihiro
Graduate School of Information Systems, The University of Electro-Communications
-
MORITA Takaya
Graduate School of Information Systems, the University of Electro-Communications
関連論文
- On Trimming Codewords of Arithmetic Codes
- Reversible Distribution Converter with Finite Precision Arithmetic
- An Information Integration Architecture for Mobile Users in WWW Environment (特集:新世代デ-タベ-ス技術--インタ-ネット・マルチメディア・モ-バイルを中心として)
- New And Robust Ray Intersection Using Interval Analysis And FAD
- SIMILARITY MEASURES FOR IMAGE RETRIEVAL BASED ON SEGMENTED REGIONS(International Workshop on Advanced Image Technology 2007)
- A New Feature for Musical Genre Classification of MPEG-4 TwinVQ Audio Data
- mm-GNAT: Index Structure for Arbitrary Lp Norm