OpenBliSSART: Design and evaluation of a research toolkit for blind source separation in audio recognition tasks

Keyphrases

• Recognition Task 100%
• Blind Source Separation 100%
• Research Toolkit 100%
• Non-negative Matrix Factorization 100%
• Audio Recognition 100%
• Algorithm Framework 40%
• Feature Extraction 20%
• Perceptual Quality 20%
• Feature Extraction Methods 20%
• Desktop Computer 20%
• Open-source Implementation 20%
• Monaural 20%
• Numerical Optimization 20%
• Source Separation 20%
• C++ Framework 20%
• Source Features 20%
• Audio Feature Extraction 20%
• Speaker Separation 20%
• Real-time Factor 20%
• Blind Audio Source Separation 20%
• Popular 20%
• Signal Recovery Algorithms 20%

Computer Science

• Blind Signal Separation 100%
• Audio Recognition 100%
• Matrix Factorization 100%
• Open Source 50%
• Feature Extraction 50%
• reconstruction algorithm 25%
• Signal Reconstruction 25%
• nonnegative matrix factorization 25%
• Audio Source Separation 25%
• Perceptual Quality 25%
• Source Separation 25%
• Multiplicity 25%