Conference proceeding
Feature analysis and selection for acoustic event detection
2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, pp 17-20
01 Jan 2008
Abstract
Speech perceptual features, such as Mel-frequency Cepstral Coefficients (MFCC), have been widely used in acoustic event detection. However, the different spectral structures between speech and acoustic events degrade the performance of the speech feature sets. We propose quantifying the discriminative capability of each feature component according to the approximated Bayesian accuracy and deriving a discriminative feature set for acoustic event detection. Compared to MFCC, feature sets derived using the proposed approaches achieve about 30% relative accuracy improvement in acoustic event detection.
Metrics
Details
- Title
- Feature analysis and selection for acoustic event detection
- Creators
- Xiaodan Zhuang - University of Illinois Urbana-ChampaignXi Zhou - University of Illinois Urbana-ChampaignThomas S. Huang - University of Illinois Urbana-ChampaignMark Hasegawa-Johnson - University of Illinois Urbana-ChampaignIEEEXiaohua Hu - Information Science (Informatics)
- Publication Details
- 2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, pp 17-20
- Series
- International Conference on Acoustics Speech and Signal Processing ICASSP
- Publisher
- IEEE
- Number of pages
- 4
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Information Science
- Web of Science ID
- WOS:000257456700005
- Scopus ID
- 2-s2.0-51449101221
- Other Identifier
- 991019167437204721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Acoustics
- Computer Science, Artificial Intelligence
- Computer Science, Cybernetics
- Engineering, Biomedical
- Engineering, Electrical & Electronic
- Imaging Science & Photographic Technology
- Mathematical & Computational Biology
- Radiology, Nuclear Medicine & Medical Imaging
- Telecommunications